Machine Learning and Knowledge Extraction doi: 10.3390/make8050130

Authors: Jorge Duarte Le Gruenwald Laurent D’Orazio Jorge Bernardino

The Index Selection Problem (ISP) refers to the task of automatically identifying the most appropriate set of indexes for a given database workload that can minimize execution costs. However, ISP is a fundamental yet complex challenge in database management systems. In the era of data-intensive applications, efficient index strategies are increasingly necessary to maintain scalability and responsiveness. This paper presents a novel automated index selection algorithm for centralized databases that employs a Double Deep Q-Network (DDQN) as the classical learning backbone and extends it with quantum-enhanced variants. Two hybrid quantum variants were proposed: Quantum Double Deep Q-Network Mixed (QDDQNM), which incorporates a residual classical pathway, and Quantum Double Deep Q-Network Boosted (QDDQNB), a boosted model without residuals. All variants were systematically evaluated using the TPC-H benchmark at two small scale factors, 10 MB and 100 MB. Experimental results show that the evaluated Deep Reinforcement Learning (DRL)-based methods improve on the SMARTIX baseline within this proof-of-concept setting. The quantum-enhanced models achieved higher best-run accumulated rewards in the reported experiments, but they also incurred substantially higher simulation cost. The results therefore suggest interesting hybrid learning behavior under the tested conditions, while also highlighting that practical scalability and cost-performance trade-offs remain important limitations for future work.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050129

Authors: Rekha R. Nair Tina Babu Balamurugan Balusamy Wee How Khoh Alaa M. Momani Basem Abu Zneid

Sepsis is a prominent cause of death in intensive care units, and delayed diagnosis greatly worsens fatal outcomes due to the complex, irregular, and uneven character of clinical time-series data. Hence we proposed an interpretable and uncertainty-aware deep learning architecture that solves data quality, temporal irregularity, and clinical explainability restrictions, which are frequently addressed separately by existing models. The suggested method combines Bidirectional Recurrent Imputation for Time Series (BRITS)-based imputation, hybrid Conditional Tabular Generative Adversarial Network-Synthetic Minority Over-sampling Technique (CTGAN-SMOTE) data augmentation, a Temporal Convolutional Networks (TCN)-Attention architecture, and continuous-time neural Ordinary Differential Equations (ODEs), along with SHapley Additive exPlanations (SHAP)-based feature attribution and uncertainty quantification. The experimental evaluation on a large ICU dataset reveals greater predictive accuracy, with an AUROC of 0.926 and accurate early warnings up to six hours before clinical onset, all while maintaining strong interpretability and calibration. The proposed framework demonstrates strong predictive performance and provides early warnings up to six hours before clinical onset, while maintaining interpretability and calibration. While the results are promising, further validation across multiple clinical settings is required to confirm its generalisability and real-world applicability.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050128

Authors: Akylbek Maxutov Nūrali Medeu Huseyin Atakan Varol

Small language models (SLMs) are valued for their computational efficiency and suitability for edge deployment, but often underperform in localized linguistic and cultural contexts due to their limited parameter size. This study explores integrating real-time web search into Qolda, a 4B-parameter Kazakh-centric SLM, to close the performance gap with larger models. We assess two strategies: Naïve Retrieval-Augmented Generation (RAG), which uses raw benchmark questions as search queries, and Query-Refined RAG, which applies various refiner models, including a supervised distillation-tuned Qolda, to optimize queries. On the KazCulture and KazMMLU benchmarks, the Naïve RAG approach in reasoning-enabled mode achieved an average accuracy of 76.00%, improving on the Zero-Shot evaluation result of 60.37%, and, in this system-level comparison, exceeding the Zero-Shot accuracy of larger open-source models such as Qwen3-32B (64.72%) and Gemma-3-27b-it (60.24%), which were evaluated without retrieval augmentation. Query refinement improved the accuracy by about 3%, from 76.00% to 79.46%, but nearly doubled the computational cost. Inference time analysis shows that Naïve RAG adds approximately 1 s of retrieval latency per question. Query refiners introduce up to 4 s of additional overhead. However, the retrieved context reduces the time required for model reasoning in think mode. The most notable gains were observed in localized cultural knowledge, where web search integration correctly answered 32.9% of KazCulture questions that the Zero-Shot baseline failed on, while losing only 16.9% in return. These results suggest that retrieval-augmented SLMs can offer a cost-effective and competitive alternative to larger models for tasks in the domains of Kazakh language and Kazakh culture.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050127

Authors: Anastasia Vangelova Adelina Aleksieva-Petrova

The growing adoption of Generative AI in education has created opportunities to automate complex pedagogical tasks, yet reliably and scalably assessing open-ended responses remains a challenge. This study proposes and evaluates an architectural solution for integrating a Large Language Model (LLM) into Moodle, combining Retrieval-Augmented Generation (RAG) and AI agent mechanisms to enable automated grading of open-ended student responses. A Moodle instance was deployed for experimental purposes, with 32 students across Bulgarian- and English-language sections, yielding data at the student (N = 32) and task (N = 160) levels, including AI-generated and instructor-assigned scores and system processing logs. The results demonstrate that the proposed system achieves substantial reductions in grading time while maintaining high agreement with expert assessments. Bias analysis revealed minimal systematic deviation across both language groups, indicating that the system preserves assessment objectivity without consistent over- or underestimation based on language. These findings suggest that a combined RAG and agentic LLM architecture can deliver efficient, accurate, and linguistically robust automated assessment within an LMS environment, offering practical design guidelines applicable to other educational platforms and similar systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050126

Authors: Patrik Viktor Gábor Kiss

Autonomous driving systems require the integration of heterogeneous sensor data, distributed V2X communication, and safety-critical decision-making into coherent and interpretable world models. This review provides a systematic analysis of knowledge graph (KG)-based approaches in autonomous driving between 2015 and 2025, following a PRISMA-aligned methodology. The literature is organised along a perception → representation → reasoning → decision taxonomy, covering traffic ontologies, V2X knowledge integration, dynamic KG updates, real-time reasoning architectures, and benchmark datasets. A clear shift from static representational ontologies toward predictive and, in a smaller subset, closed-loop validated neuro-symbolic architectures. Knowledge graphs emerge as semantic integration layers that improve contextual reasoning, explainability, and rule compliance in safety-critical environments. Key challenges include scalable real-time reasoning, standardised evaluation frameworks, and safety-aligned integration of learning-based components.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050125

Authors: Tymoteusz Miller

Artificial intelligence systems deployed in safety-critical and regulated environments require not only predictive performance, but also strict adherence to operational constraints, auditability, and reproducibility. However, in most contemporary architectures, governance is treated as an external or post hoc mechanism, limiting the ability to ensure consistent and verifiable decision execution. This paper introduces Controlled Agentic AI Systems (CAIS), a formal architectural framework in which governance is embedded directly into the decision pipeline as a deterministic operator. The proposed formulation integrates a decision model, a constraint specification, and a governance operator that transforms proposed actions into admissible executed actions. The framework further defines audit trace semantics and replayability conditions, enabling deterministic reconstruction of decision trajectories. Theoretical analysis demonstrates that, under standard regularity assumptions, governance can be modeled as a non-expansive projection that enforces constraint-aware decision transformation while inducing bounded decision drift. This provides formal guarantees that governance does not destabilize system dynamics under perturbations. To evaluate these properties, we implement a reference CAIS architecture and conduct controlled experiments in multi-agent and federated simulation environments. The results show that embedding governance significantly reduces the frequency and severity of constraint violations across a range of scenarios. Projection-based repair consistently outperforms approval-only strategies, achieving near-complete compliance in structured regimes while maintaining bounded intervention costs. Importantly, governance does not degrade stability or convergence in federated settings and, in some cases, reduces action-level variance induced by distributed training. While strict feasibility cannot be guaranteed in all practical settings due to approximation and solver limitations, the empirical findings confirm that governance acts as a stabilizing transformation that consistently improves compliance without introducing destabilizing effects. The CAIS framework establishes governance as a first-class architectural component of agentic AI systems, providing a unified foundation for designing constraint-aware, auditable, and reproducible decision pipelines in regulated environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050124

Authors: Fatemeh Tabatabaei Hamzeh Khalili

This paper presents a predictive analytics framework for dynamic resource allocation in next-generation networks, with a focus on 5G and 6G scenarios. The proposed approach uses Long Short-Term Memory (LSTM) models to predict traffic patterns and proactively support automated resource management decisions in network environments. Building on our previous work, we introduce a selective knowledge transfer mechanism, termed “cognitive transfer,” which allows for the reuse of relevant components from previously trained models. This method consolidates multiple models into a compact, generalized form and transfers only the most relevant segments to new traffic scenarios, significantly reducing the need for training from scratch, especially when local data is limited. This maintains an initial consolidated (base) model that is adapted to new traffic scenarios by selecting the corresponding segment from the most relevant parts of the model, identified via a scoring metric. To complement this, a decision-based model retraining strategy is integrated to monitor prediction accuracy, triggering updates only when performance degrades to reduce computational overhead. The framework is evaluated using the Abilene network topology and traffic dataset. Results demonstrate that the approach maintains high prediction accuracy while minimizing under-provisioning, which is critical for avoiding packet loss and ensuring service continuity in high-reliability applications. Across four network links, our method reduced the average number of re-trainings by 1.8× under relaxed thresholds compared to full model transfer, while the prediction error increased by a negligible 0.0008.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050123

Authors: Zaharias D. Zaharis Emmanouil Georgios Nikopolitidis Pavlos I. Lazaridis Panagiotis Sarigiannidis Sotirios K. Goudos

First introduced in 1995, the Particle Swarm Optimization (PSO) algorithm offers a reliable and efficient solution to real-valued optimization problems. However, extending it to binary-valued problems proved challenging. This paper proposes a new Boolean version of the PSO technique based on a novel mutation strategy. By employing an innovative mutation mechanism in the velocity bitstring, the method enforces a minimum perturbation level, reduces the risk of premature convergence, and promotes broader global search. Several variations of the algorithm and parameter combinations are evaluated using 47 benchmark functions to derive the best-performing configuration, which is then compared with other population-based methods to demonstrate the effectiveness of the proposed algorithm. Finally, the technique is applied to an antenna array thinning problem for the design of a planar antenna array with certain specifications.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050122

Authors: Daniil Pasenko Roman Davydov

Deformable medical image registration aims to spatially align anatomical structures across volumetric scans. Recent transformer-based methods achieve high overlap accuracy but often produce deformation fields with topological violations. We propose CTCF, a Cascade Transformer for Coarse-to-Fine registration that wraps a lightweight coarse-and-refined envelope around a core registration module. Level 1 provides a coarse displacement estimate at quarter resolution, Level 2 performs the main registration via a Swin Transformer encoder with deformable cross-attention and a learned super-resolution decoder, and Level 3 applies error-driven flow refinement at half resolution. The two outer levels add only 3.0% parameter overhead yet improve registration accuracy while maintaining competitive deformation regularity relative to external baselines. The model is trained end-to-end with a composite unsupervised loss combining local normalized cross-correlation, diffusion regularization, inverse-consistency, and Jacobian-based topology preservation. On the OASIS brain MRI benchmark, CTCF achieves the highest Dice score of 0.8208 among the compared unsupervised methods while maintaining competitive SDlogJ, with all Dice improvements statistically significant at p<0.001 by the Wilcoxon signed-rank test. On IXI, CTCF also achieves the best Dice, HD95, SDlogJ, and fold percentage among the compared methods. A five-round ablation study validates each component: cascade decomposition isolates each level’s contribution, and resolution scaling experiments confirm the framework’s scalability, yielding further accuracy gains with zero parameter overhead.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050121

Authors: Josue Genaro Almaraz-Rivera Jose Antonio Cantoral-Ceballos Juan Felipe Botero

Self-Supervised Learning (S-SL) is a recent line of research that could represent the next step to understanding human intuition. By blending the strengths of unsupervised and supervised learning paradigms, S-SL endows Deep Learning models with stronger generalization capabilities. Although better known for its applications in Computer Vision and Natural Language Processing, S-SL has also proved its value in other fields, such as cybersecurity. In this work, we review the current progress and future trends in S-SL for the two most relevant problems discussed in the cybersecurity literature: network intrusion and malware detection. The scope of this survey spans from 2019 to 2025. From an initial analysis of over 200 documents, we distill the 50 most relevant papers. We also highlight opportunity areas, such as attack detection over encrypted network traffic, RAM-based analysis of obfuscated malware, creating S-SL models for tabular data and resource-constrained devices, as well as the research on backdooring, encoder extraction, the transferability of vulnerabilities, and data memorization in S-SL. To the best of our knowledge, this is the first comprehensive survey regarding the application of Self-Supervised Learning in cybersecurity, benchmarking contrastive learning vs auxiliary pretext tasks and presenting the data requirements for implementing S-SL solutions in this field. We hope this paper provides a firm ground for further exploration.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050119

Authors: Helala AlShehri Mariam Busaleh

Robust brain tumor classification from magnetic resonance imaging (MRI) remains challenging due to complex structural heterogeneity and subtle inter-class variability. Beyond predictive accuracy, conventional convolutional neural networks predominantly rely on texture-dominant features and fixed receptive fields, which may limit the extraction of clinically meaningful structural information. This study proposes a morphology-aware multi-scale deep representation learning framework that embeds morphological inductive bias directly within hierarchical feature extraction. The proposed architecture synergistically integrates trainable morphological operations with multi-scale convolutional feature learning inside a unified residual framework, supported by an in-block morphological refinement mechanism and a morphology-aware downsampling module. Unlike prior approaches that treat morphological operators as preprocessing or auxiliary branches, the proposed design incorporates differentiable dilation and erosion into the core feature hierarchy to guide structure-aware representation formation. The model was evaluated using five-fold cross-validation and an independent test set, achieving an overall test accuracy of 99.31% with consistently high macro-averaged precision, recall, F1-score, and AUC values. Grad-CAM analysis further demonstrates that the learned representations emphasize clinically relevant tumor regions, supporting interpretable structural knowledge extraction. Ablation studies confirm that performance improvements arise from the synergistic integration of multi-scale learning and morphology-aware refinement. Overall, embedding structural inductive bias within multi-scale deep representation learning enhances robustness, stability, and interpretable knowledge extraction for brain tumor MRI analysis.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050120

Authors: Vassiliy Serbin Mateus Mendes Aray Kassenkhan Akbayan Bekarystankyzy Gulnur Ibragim Azamat Tolegenov

The rapid digitalization of higher education has intensified the need for reliable methods to assess higher-order cognitive skills, particularly critical thinking, in adaptive learning environments. However, most existing assessment approaches rely primarily on test outcomes and academic performance indicators, which do not adequately capture the multidimensional and process-based nature of critical thinking. This study proposes a multi-criteria hierarchical model for identifying and quantitatively assessing students’ critical thinking based on individual process indicators of learning activity in an intelligent educational environment. The model integrates cognitive, metacognitive, and behavioral indicators, including knowledge dynamics, task complexity, time characteristics, learning activity intensity, error rate, level of doubt, user interaction patterns, and system operating modes. These indicators are aggregated into a three-component structure representing metacognitive awareness, analytical depth, and strategic learning activity. The proposed model was empirically validated through a quasi-experimental longitudinal study involving 500 university students divided into control and experimental groups. The results demonstrate a statistically significant increase in all latent components of critical thinking and in the integral indicator within the experimental group. The model shows satisfactory internal consistency (Cronbach’s α≥0.77) and acceptable construct validity confirmed by confirmatory factor analysis. The findings indicate that the proposed model can serve as a practical analytical tool for monitoring critical thinking development and supporting personalized learning trajectories in adaptive digital educational systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050118

Authors: Elif Fırıncı Mohanad Alayedi

The appearance of technical debt (TD) becomes a critical problem, posing challenges related to software maintainability and its quality within the context of fast modern software development. The presented research focuses on the issue of TD appearance in the context of Python software development by employing a hybrid approach involving perception and predictive approaches. Within the scope of research, the perceptions of 86 IT practitioners and developers have been studied with regard to their reactions, adaptation, and prioritization of different types of TDs. According to the qualitative results, cyclic architectural debt stems from low test coverage, documentation deficiency, and complicated code structure. Based on the aforementioned information, the research team developed a dataset consisting of 130 Python codes in real conditions with the following characteristics: code complexity, comments-to-code ratio, code smells, and software maintainability indexes being used. Thereafter, the application of DT, LR, NB, SVM, KNN, and RF predictive models allowed detecting TDs. The presented results reveal the possibility of predicting TDs with the use of machine learning methods, with optimal performance provided by random forest and optimized logistic regression models.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050117

Authors: Vitaliy Vorobyov Leonid Ugadiarov Vladimir Frolov Alexey Kovalev Aleksandr Panov

In this paper, we propose a novel approach, COMPAS (COMPose Actions and Slots), which leverages the strengths of state-of-the-art object-centric approaches for modeling the dynamics of an environment. Our method encodes the environment’s state into symbol-like, object-centric representations, known as slots, where each slot corresponds to an individual object. This approach offers a structured and interpretable way to model complex environments by combining slots with action representations for accurate next-state prediction. The primary contribution of our work is an efficient world model with a dynamics predictor capable of predicting accurate trajectories in action-dependent environments. Additionally, our slot extractor module enhances the predictive capabilities by extracting deterministic slots that remain consistent both within a single trajectory and across episodes. Unlike slots sampled from a trainable distribution, deterministic slots are generated from a single trainable parameter together with slot positional embeddings. This design improves the consistency across episodes, which in turn leads to more accurate dynamics prediction. We present a comprehensive evaluation of our approach in various environments, demonstrating that our proposed method outperforms competing models in environments with discrete and continuous action spaces.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050116

Authors: Chaymae Yahyati Ismail Lamaakal Yassine Maleh Khalid El Makkaoui Ibrahim Ouahbi

Online and blended classrooms widen access but remove the in-person cues instructors use to gauge attention. Prior work typically relies on heavy, cloud-bound or multimodal models that are hard to deploy on commodity laptops, treats attention as an unordered label without calibrated probabilities, and evaluates on subject-overlapping splits with limited robustness analysis. This creates a gap in Tiny, deployable, calibration-aware methods validated under realistic protocols. We address this gap with a TinyML, vision-only pipeline that estimates four attention levels: (Very Low, low, high, Very High ) from short webcam clips under strict on-device budgets. Each clip of T=30 frames at 224×224 is processed by a compact hybrid encoder: a CNN extracts per frame spatial features, a BiLSTM models temporal context, and a lightweight GRU refines dynamics; three parallel branches with staggered widths encourage feature diversity before fusion. We apply structured pruning of convolutional channels and recurrent units, post-training INT8 quantization, and temperature scaling for calibrated probabilities; models are exported as ONNX. On DAiSEE with subject-independent splits, the baseline attains 99.86% accuracy and 0.998 macro-F1, with strong ordinal agreement (QWK = 0.998, ordinal MAE = 0.03). The compressed model preserves reliability (macro-F1 = 0.995, QWK = 0.995), remains robust to low light, partial occlusion, and head yaw, and yields ∼4× smaller size and ∼2.3× CPU speedups. These results indicate a deployable, privacy-preserving approach to fine-grained, on-device attention analytics.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050115

Authors: Kinga Lasek Michal Ptaszynski Fumito Masui Mujahid Khalifah Juuso Eronen

Despite the progressing standardization of the Japanese language, regional dialects persist, particularly among older generations, causing communication gaps, which results in problems especially in healthcare and emergency contexts. This study proposes a text-to-text normalization method to convert eight Japanese dialects into standard Japanese using a fine-tuned mT5-small architecture. We evaluate the impact of learning rate schedulers, training duration, and data preprocessing on model performance. Our results demonstrate that the CharacTER (Character Translation Edit Rate) metric provides a more accurate evaluation than BLEU, which is practically ill-suited for the unsegmented nature of Japanese text. The optimal configuration minimizes character error rates by aligning input data with natural, unspaced Japanese orthography. Furthermore, we observe a statistically significant correlation between the model’s conversion error rate and the physical distance of the source dialect from Tokyo. This finding suggests that the model’s performance effectively serves as a proxy for measuring linguistic distance between dialectal variations and the standard language.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050114

Authors: Iulia Onică Iulian B. Ciocoiu

The paper addresses ECG arrhythmia classification using a novel input fusion strategy that combines spatial representations of ECG time series recordings. Four distinct time series-to-image transformations are considered, namely classical spectrograms, Gramian Angular Field (GAF), Recursive Plot (RP), and the S-Transform (ST). Classification of combined 2 × 2 images generated from single-lead ECG recordings is performed using both custom and ResNet-50 deep learning architectures. Finally, several distinct explainability algorithms are used to identify the relevant regions in the input images that mainly influence the classification decisions. Experiments performed on the MIT-BIH and Chapman–Shaoxing arrhythmia datasets revealed performance comparable to more sophisticated approaches in terms of accuracy (99%), F1-score (98.6%), and AUC (0.999) values.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050113

Authors: Ngoc Tuyen Do Minh Nguyen Quang Hai Van Pham

Structured radiology reporting can improve clinical decision support by standardizing clinical findings into hierarchical formats. However, thousands of questions in structured report templates about clinical findings are prohibitively time-consuming, which can limit clinical adoption. Furthermore, early medical VQA datasets primarily focused on free-text and independent question–answer pairs while a recent dataset, Rad-ReStruct, introduced a hierarchical VQA, but the accompanying model still relies heavily on flattened embedding representations and single-path text–image fusion mechanisms that inadequately handle complex hierarchical dependencies in responses. In this paper, we propose DPA-HiVQA (Dual-Path Cross-Attention for Hierarchical VQA), addressing these limitations through two key contributions: (1) multi-scale image embedding representing global semantic embeddings with patch-level spatial features from domain-specific BioViL encoder; (2) dual-path cross-attention mechanism enabling simultaneous holistic semantic understanding and fine-grained spatial reasoning. Evaluated on the Rad-ReStruct benchmark, the model substantially outperforms the established benchmark baseline with an overall F1-score and Level 3 F1-score improvement by 21.2% and 31.9%, respectively. The proposed model demonstrates that dual-path cross-attention architectures can effectively connect holistic semantic understanding and fine-grained spatial detail, paving the way for practical AI-assisted structured reporting systems that reduce radiologist burden while maintaining diagnostic accuracy.

Machine Learning and Knowledge Extraction doi: 10.3390/make8050112

Authors: Mohammed Alnemari Rizwan Qureshi Nader Bagherzadeh

Neural scaling laws describe how model performance improves as a power law with size, but existing work has focused almost entirely on models above 100 M parameters. The regime below 20 million parameters, where TinyML and edge AI systems operate, remains largely unexamined. We train 90 models spanning 22 K to 19.8 M parameters across two architecture families (a plain ConvNet and MobileNetV2) on CIFAR-100, varying width while holding depth and training protocol fixed. Both architectures follow approximate power laws, with exponents of α=0.156 (ScaleCNN) and α=0.106 (MobileNetV2). However, the power law does not hold uniformly: local exponents decay with scale, and MobileNetV2 saturates at 19.8 M parameters (αlocal=0.006), hitting a data wall. The structure of errors also changes with scale. The Jaccard overlap between error sets of the smallest and largest ScaleCNN models is only 0.35; compression changes which inputs are misclassified, not merely how many. Small models develop a triage strategy, concentrating capacity on easy classes (Gini of per-class accuracy: 0.26 at 22 K params vs. 0.09 at 4.7 M) while effectively abandoning the hardest ones (bottom-5 class accuracy: 10% vs. 53%). The smallest models achieve the lowest ECE values (0.013 vs. peak 0.110 at mid-size), reversing the typical overconfidence–capacity relationship, though this partly reflects a global-mean matching artifact rather than well-calibrated per-bin confidence. On CIFAR-100, aggregate accuracy alone is therefore a misleading basis for edge deployment decisions; validation must happen at the target model size. All findings in this study are based on CIFAR-100 (32 × 32, 100 classes); their generalizability to other datasets, resolutions, and architectures remains to be verified.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040111

Authors: Zine. Eddine Louriga Ismail Jabri Aziza El Ouaazizi Anass El Affar

Iris biometric systems are recognized as secure alternatives to conventional authentication methods, yet challenges such as variable illumination, noise, and intricate iris textures persist. To address these issues, our study presents a novel hybrid iris recognition framework that integrates advanced deep learning with a pioneering application of Multivariate Ensemble Empirical Mode Decomposition (MEEMD) for feature extraction—a method not previously applied in this context. Our framework first employs MEEMD to extract statistical features that capture the iris’s nonlinear and nonstationary variations. We then combine global semantic information from two pretrained convolutional neural networks—VGG16 and ResNet-152—with local micro-texture details encoded by Local Binary Patterns (LBP) to form a comprehensive feature representation. An efficient pre-processing and segmentation stage precisely isolates the iris region, and the resulting features are refined through dimensionality reduction techniques to yield a robust, compact representation. These features are subsequently classified using multiple models, each rigorously tuned via hyperparameter optimization. Experimental validation on benchmark datasets—including IITD, CASIA, and UBIRIS.v2—shows that our model achieves recognition rates of up to 98% on IITD, 97% on CASIA, and 97.30% on UBIRIS.v2, surpassing existing approaches. This work not only enhances iris recognition performance but also establishes a novel method that bridges advanced deep learning with innovative feature extraction for high-security applications.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040110

Authors: Ahmed Ibrahim Abdullah Hosseini Hoda Helmy Maryam Arabi Aya AlShareef Wafa Lakhdhar Ahmed Serag

Dialectal variation presents a major challenge for deploying medical language models in real-world healthcare settings, where patient–clinician communication often occurs in regional vernaculars rather than standardized language forms. This challenge is particularly pronounced in the Arabic-speaking world, where clinical interactions frequently take place in diverse dialects that differ substantially from Modern Standard Arabic. Fine-tuning and maintaining separate models for each dialect is computationally inefficient and difficult to scale, motivating more integrated approaches. In this work, we present MENARA, an Arabic medical language model constructed by merging Egyptian Arabic, Moroccan Darija, and medical-domain specialists through model merging. We extend prior feasibility findings through comprehensive evaluation of cross-dialect performance, medical safety, and cross-lingual knowledge retention. Specifically, we introduce a fine-grained dialect composition analysis to quantify lexical purity and structured code-switching behavior, benchmark against state-of-the-art Arabic LLMs, conduct subject-matter-expert assessment of both dialectal fidelity and medical appropriateness. The results show that model merging preserves core medical competence while enabling robust dialectal adaptation, achieving strong cross-dialect fidelity while substantially reducing storage and deployment overhead compared to maintaining separate models. These findings establish model merging as a potentially practical and resource-efficient paradigm for dialect-aware medical NLP in linguistically fragmented healthcare environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040109

Authors: Neelabh Kshetry Mehmed Kantardzic

Counterfactual explanations are increasingly vital for understanding and trusting machine learning models. This paper presents Desirability Rating-based Counterfactual (DeRaC), which is a generalized framework for generating valid counterfactual explanations applicable to classification problems with complex output spaces, including single and multi-output classification with binary and multi-class outputs. By expanding the definition of counterfactual validity through a novel “desirability rating,” the approach addresses limitations in existing methods regarding complex output spaces. The framework introduces concepts such as partially valid counterfactuals and a quantitative measure of output desirability. It can be integrated with various objective functions to identify counterfactuals that satisfy properties such as similarity, proximity, and validity. Experiments demonstrate the feasibility of systematically generating counterfactuals using existing optimization techniques, achieving varying degrees of validity and similarity; specifically, Genetic Algorithm produces consistently higher counterfactual desirability albeit at the expense of longer computation times. We observed a higher average counterfactual desirability rating of 0.871 across all tested optimization methods with Powell’s method combined with DeRaC achieving the lowest average distance of 0.897 when using a mixed-objective function. The research emphasizes the context-dependent nature of counterfactuals and lays the foundation for more transparent and trustworthy machine learning systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040108

Authors: Mateus Mendes Torres Farinha

As modern computers became increasingly more popular and larger amounts of digital data were available, different methodologies were proposed to extract information from data. CRISP-DM methodology quickly spread and is currently one of the most popular approaches used for data analysis. However, it has some shortcomings, such as being too general or business-centered. Different authors have proposed variations more suitable to specific fields in order to overcome those limitations. The present paper reviews CRISP-DM, some variations and similar methodologies, and proposes a Methodology for Industrial Data Analysis (MIDA)—a methodology conceived and improved over time, based on previous experience in industrial engineering processes. MIDA consists of eight steps and partially overlaps with CRISP-DM. It has been successfully applied in several previous projects.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040106

Authors: Nuno Ferreira Filipe Pinto António Valente Diana Augusto Manuela Reis Salviano Soares

Old-vine vineyards often contain dozens of grapevine varieties intermingled and irregularly distributed, making plant-level varietal identification slow and expensive when based on ampelography or molecular approaches. This paper proposes a field-oriented computer-vision pipeline for Vitis vinifera variety identification using images with a natural background from the historic “Vinha Maria Teresa” parcel (Quinta do Crasto, Portugal). A single-class YOLO11 detector is trained to localize the vine leaf and generate standardized crops, and a YOLO11 classifier is then fine-tuned on leaf regions of interest (ROIs) for eight selected varieties in the Douro UNESCO region. We annotated 2015 vineyard images for classification and supplemented detection training with 2648 additional leaf images; detectors (YOLO11n/s/m) were benchmarked under four augmentation regimes and evaluated on a fixed 48-image subset, including runtime on CPU and GPU. The best detector reached mAP@50–95 of 0.918 on the benchmark, while YOLO11n achieved ∼27 FPS on CPU for fast cropping. On a 303-image test set, the best classifier (YOLO11s with mixed augmentations) achieved 94.06% Top-1 accuracy, 93.92% macro-F1, and 100% Top-5 accuracy with remaining errors concentrated among morphologically similar varieties. To assess deployment-oriented performance, classifiers trained under three input settings (manual crops, detector-generated crops, and full images) were evaluated on a held-out 48-image benchmark subset; removing the detection step reduced Top-1 accuracy from 75.00% to 68.75%, while the gap between manual and automatic crops was only 2.44 pp on successfully detected images with detection failures (14.6%) representing the primary operational bottleneck. Repeated retraining of the best manual-crop YOLO11s configuration across multiple random seeds showed stable performance with low variability in Top-1 accuracy and macro-F1. Under identical training conditions, ResNet50 and EfficientNet-B0 provided competitive baselines, but YOLO11s remained the strongest overall model on the held-out field benchmark. These results indicate that lightweight leaf detection plus crop-based classification can support scalable varietal identification in old vineyards under realistic acquisition conditions.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040107

Authors: Sakorn Mekruksavanich Anuchit Jitpattanakul

The rapid growth of the elderly population worldwide demands reliable activity recognition technologies to support independent living and continuous health supervision. However, conventional wearable sensor-based human activity recognition (HAR) techniques often fail to capture the complex temporal behaviour and subtle motion patterns characteristic of the elderly. To address these limitations, this study introduces a hybrid deep residual architecture—CNN-CBAM-BiGRU—that integrates convolutional neural networks (CNNs), the convolutional block attention module (CBAM), and bidirectional gated recurrent units (BiGRUs) to improve activity recognition using inertial measurement unit (IMU) data. In the proposed CNN-CBAM-BiGRU framework, CNN layers automatically derive representative features from raw sensor signals, CBAM applies adaptive channel and spatial attention to highlight informative patterns, and BiGRU captures long-range temporal relationships within activity sequences. The approach was evaluated on three benchmark datasets designed for elderly populations—HAR70+, HARTH, and SisFall—covering daily activities and fall events. The proposed model consistently outperforms existing methods across all datasets, achieving accuracies exceeding 96%, F1-scores above 93%, and a fall detection recall of 93.74%, confirming its robustness and suitability for safety-critical monitoring applications. Class-level evaluation indicates excellent recognition of static postures and consistent performance for dynamic actions. Convergence analysis further confirms efficient learning with limited overfitting across datasets. The proposed framework thus provides a robust and accurate solution for wearable-based elderly activity recognition, with strong potential for deployment in fall detection, health monitoring, and ambient assisted living systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040105

Authors: Hiba Adil Al-kharsan Róbert Rajkó

Brain tumor classification from magnetic resonance imaging, which is also known as MRI, plays a sensitive role in computer-assisted diagnosis systems. In recent years, deep learning models have achieved high classification accuracy. However, their sensitivity to adversarial perturbations has become an important reliability concern in medical applications. This study suggests a robust brain tumor classification framework that combines non-negative matrix factorization (NNMF or NMF), lightweight convolutional neural networks (CNNs), and diffusion-based feature purification. Initially, MRI images are preprocessed and converted into a non-negative data matrix, from which compact and interpretable NNMF feature representations are extracted. Statistical metrics, including AUC, Cohen’s d, and p-values, are used to rank and choose the most discriminative components. Then, a lightweight CNN classifier is trained directly on the selected feature groups. To improve adversarial robustness, a diffusion-based feature-space purification module is introduced. A forward noise method followed by a learned denoiser network is used before classification. System performance is estimated using both clean accuracy and robust accuracy under powerful adversarial attacks created by AutoAttack. The experimental results show that the proposed framework achieves competitive classification performance while significantly enhancing robustness against adversarial perturbations. The findings presuppose that combining interpretable NNMF-based representations with a lightweight deep approach and diffusion-based defense technique supplies an effective and reliable solution for medical image classification under adversarial conditions.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040104

Authors: Maria C. Mariani Sourav Malakar Amrita Bagchi Subhrajyoti Basu Saptarsi Goswami Osei Kofi Tweneboah Sarbadeep Biswas Ankit Dey Ankit Sinha

This study provides a comparative evaluation of three state-of-the-art large language models (LLMs), namely OpenAI’s (San Francisco, CA, USA) GPT-4.0, Google’s (Google LLC, Mountain View, CA, USA) Gemini 2.0 Flash, and Meta’s (Meta Platforms, Menlo Park, CA, USA) LLaMA-4-Scout-17B-16E, in a decision-oriented framework in which the models generate structured outputs based only on historical closing-price data. The evaluation covers 150 stocks sampled from three countries (India, the United States, and South Africa) across ten economic sectors, including Information Technology, Banking, and Pharmaceuticals. Unlike many prior studies that combine numerical and textual inputs, this study relies solely on three years of numerical time series data and examines model responses in terms of decision labels such as buy, sell, or hold. The LLMs were provided with historical closing-price sequences and prompted with three types of finance-related questions: (a) whether to buy a stock, (b) whether to sell or hold a stock, and (c) in a pairwise comparison, which stock to buy or hold. These prompts were evaluated across two investment horizons: 1 month and 3 months. Model outputs were compared against realized market outcomes during the corresponding test periods. Performance was assessed across four key dimensions: country, sector, annualized volatility, and question type. The models were not given any supplementary financial information or instructions on specific analytical methods. The results indicate that GPT-4.0 achieves the highest average accuracy (56%), followed by LLaMA-4-Scout-17B-16E (48%) and Gemini 2.0 Flash (39%). Overall performance remains moderate and varies across market conditions, with relatively higher accuracy observed in high-volatility regimes (51%). This work evaluates how LLMs behave when presented with structured numerical price sequences in a controlled decision-labeling setting and contributes to the broader discussion on the potential and limitations of LLMs for numerical decision tasks in finance.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040103

Authors: Abdul Sittar Mateja Smiljanic Alenka Guček Marko Grobelnik

The proliferation of fake news across social media, headlines, and news articles poses major challenges for automated detection, particularly in multilingual and cross-media settings affected by data imbalance. We propose a fake news detection framework based on LLM-driven, feature-guided text augmentation. The method generates realistic synthetic samples across languages, media types, and text granularities while preserving meaning and stylistic coherence. Experiments with classical and transformer-based models (Random Forest, Logistic Regression, BERT, XLM-R) across social media, headlines, and multilingual news datasets show consistent improvements in performance. For inherently balanced datasets (e.g., social media), synthetic augmentation yields negligible but stable performance changes. Across imbalanced scenarios, synthetic augmentation substantially improves minority-class recall and F1-score (e.g., fake news recall from 0.57 to 0.86), while preserving majority-class performance, leading to more balanced and reliable classifiers, whereas oversampling significantly degrades results due to overfitting on duplicated language patterns. Overall, a hybrid semantic- and style-based model proves to be the most robust strategy, outperforming oversampling and matching or exceeding baseline performance across datasets.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040102

Authors: Saim Rasheed

Automated face mask detection remains an important component of hygiene compliance, occupational safety, and public health monitoring, even in post-pandemic environments where real-time and non-intrusive surveillance is required. Traditional deep learning models provide strong recognition performance but are often impractical for deployment on embedded and edge devices due to their computational and energy demands. Recent research has therefore emphasized lightweight and hybrid architectures that seek to preserve detection accuracy while reducing model complexity, inference latency, and power consumption. This review presents an architecture-centered synthesis of face mask detection systems, examining conventional convolutional models, lightweight convolutional networks such as the MobileNet family, and hybrid frameworks that integrate efficient backbones with optimized detection heads. Comparative analysis of reported results highlights key trade-offs between accuracy, efficiency, and deployment feasibility under heterogeneous datasets, evaluation protocols, and hardware settings. Open challenges, including improper mask detection, domain adaptation, model compression, and the extension of mask detection toward broader Personal Protective Equipment (PPE) compliance monitoring, are discussed to outline a forward-looking research agenda. Overall, this review consolidates current understanding of architectural design strategies for face mask detection and provides guidance for developing scalable, robust, and real-time deep learning solutions suitable for embedded and mobile platforms.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040101

Authors: Dominik Polke Elmar Ahle Dirk Söffker

Gaussian processes (GPs) are a popular method in machine learning (ML) to model complex systems. One advantage of GPs over other ML models is their ability to quantify uncertainty in predictions. In the past, many advanced methods for GPs have been developed and published for various applications. Adaptive learning (ADL) is one of these applications, in which the consideration of uncertainty prediction plays a major role. The goal of ADL is to replace costly and time-consuming experiments and simulations of complex systems with surrogate models. This is achieved by strategically minimizing queries to maximize efficiency. In the ML literature, various reviews cover either GP methods or ADL strategies. Their focus is more on specific aspects. A comprehensive overview of different GP methods in various ADL applications was missing. This review categorizes GPs and related advanced methods for the first time in the context of ADL applications. A classification is provided for advanced GP methods, ADL methodologies, and practical application areas of GPs with ADL. This review distinguishes between ADL strategies with single-point and batch-query methods for Bayesian optimization and active learning, and highlights real-world applications such as material and product design, as well as efficient modeling for costly simulations and experiments. By combining these aspects, it offers a comprehensive guide for researchers and practitioners applying ADL with GPs to their specific use cases.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040100

Authors: Hafida Hidani Safâa El Ouahabi Mouncef Filali Bouami

The rapid growth of Arabic social media content requires the development of accurate and efficient methods for sentiment analysis. We propose a resource-efficient multi-task learning (MTL) framework for modern standard Arabic (MSA). The model uses a shared AraBERT encoder to jointly predict emotion, polarity, and intention. We integrate knowledge distillation (KD) from a large teacher model, self-distillation (SD) using model self-ensembling, and adversarial training (AT) as a regularization strategy. Experiments conducted on an annotated corpus of MSA tweets demonstrate that all distilled models outperform a fine-tuned multi-task baseline, and the combined KD+SD+AT configuration achieves competitive results. For instance, KD alone raised Macro F1 for emotion from 0.83 to 0.88 and for intention from 0.67 to 0.72. KD+SD+AT achieved the best intention F1 (0.76) and the highest polarity F1 (0.90). Notably, F1-scores for several minority classes show consistent improvement, particularly under KD and combined configurations. Paired t-tests confirm that several improvements, especially those obtained with KD and KD+SD+AT, are statistically significant (p<0.05). Our results indicate that distillation, combined with adversarial regularization, enables the development of smaller and more efficient Arabic sentiment models while maintaining competitive accuracy. These findings address a gap in Arabic multi-task sentiment analysis and provide a scalable, resource-efficient framework, along with empirical insights for distillation in Arabic language models.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040099

Authors: Rohan Le Roux Siavash Khaksar Mohammadali Sepehri Iain Murray

Open-pit mining operations rely heavily on visual inspection to identify indicators of slope instability such as surface cracks. Early identification of these geotechnical hazards enables timely safety interventions to protect both workers and assets in the event of slope failures or landslides. While computer vision (CV) approaches offer a promising avenue for autonomous crack detection, their effectiveness remains constrained by the scarcity of labelled geotechnical datasets. Deep learning (DL)-based models, in particular, require large amounts of representative training data to generalize to unseen conditions; however, collecting such data from operational mine sites is limited by safety, cost, and data confidentiality constraints. To address this challenge, this study proposes a novel hybrid game engine–generative artificial intelligence (AI) framework for large-scale dataset generation without requiring real-world training data. Leveraging a parameterized virtual environment developed in Unreal Engine 5 (UE5), the framework generates realistic images of open-pit surface cracks and enhances their fidelity and diversity using StyleGAN2-ADA. The synthesized datasets were used to train the YOLOv11 real-time object detection model and evaluated on a held-out real-world dataset of open-pit slope imagery to assess the effectiveness of the proposed framework in improving model generalizability under extreme data scarcity. Experimental results demonstrated that models trained using the proposed framework consistently outperformed the UE5 baseline, with average precision (AP) at intersection over union (IoU) thresholds of 0.5 and [0.5:0.95] increasing from 0.792 to 0.922 (+16.4%) and 0.536 to 0.722 (+34.7%), respectively, across the best-performing configurations. These findings demonstrate the effectiveness of hybrid generative AI frameworks in mitigating data scarcity in CV applications and supporting the development of scalable automated slope monitoring systems for improved worker safety and operational efficiency in open-pit mining.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040098

Authors: Sarthak Pattnaik Chhayank Jain Eugene Pinsky

This study illuminates fundamental questions in behavioral science through advanced machine learning methodologies applied to large-scale public opinion data. Drawing on Kahneman and Tversky’s dual-process theory and Sunstein’s nudge architecture, we employ hierarchical unsupervised clustering and supervised predictive models to detect cognitive biases—loss aversion, availability heuristic, and partisan motivated reasoning—embedded within a nationally representative survey of 5022 American respondents. Our primary methodological contribution is a hierarchical two-stage clustering framework that uncovers latent opinion structures without imposing a priori partisan categories, permitting discovery of cross-cutting cleavages invisible to conventional survey analysis. Three principal findings emerge: (1) loss aversion is empirically confirmed in prospective economic perception, with pessimists outnumbering optimists at a 1.14:1 ratio even among respondents rating current conditions positively; (2) partisan motivated reasoning produces a 13.15 percentage-point perception gap among individuals with identical financial circumstances; and (3) multi-platform digital engagement is associated with reduced partisan bias, providing evidence that challenges simple echo chamber assumptions. Crime safety perception emerges as the strongest predictor of economic bias, surpassing party affiliation, and substantiating availability heuristic dominance in political cognition. These findings carry implications for democratic accountability, platform governance, and the ethics of AI-augmented behavioral analysis in an era of affective polarization.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040097

Authors: Antonio Jesús Banegas-Luna Horacio Pérez-Sánchez Carlos Martínez-Cortés

While predictive accuracy is often prioritized in machine learning (ML) models, interpretability remains essential in scientific and high-stakes domains. However, diverse interpretability algorithms frequently yield conflicting explanations, highlighting the need for consensus to harmonize results. In this study, six ML models were trained on six synthetic datasets with known ground truths, utilizing various model-agnostic interpretability techniques, as well as gradient-based and counterfactual-based explainers. Consensus explanations were generated using established methods and a novel approach: WISCA (Weighted Scaled Consensus Attributions), which integrates class probability and normalized attributions. WISCA consistently aligned with the most reliable individual method, underscoring the value of robust consensus strategies in improving explanation reliability.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040096

Authors: Sarala Ghimire Turgay Celik Martin Gerdes Christian W. Omlin

Data-driven approaches to blood glucose predictive modeling face significant challenges due to the inherent variability in biological systems. While these methods efficiently capture statistical patterns through automated processes, they often lack the biological interpretability necessary to link model behavior with underlying physiological mechanisms. In contrast, physiological models offer accurate mechanistic representations but require complex parameterization and specialized domain expertise. In this work, we present an approach for predicting blood glucose levels (BGLs) leveraging the concept of physiology-informed neural networks (PINNs). This approach addresses the challenge of BGL prediction by incorporating the parameters of insulin and meal dynamics within the architecture of a predictive network. It employs a two-stage learning approach for modeling physiology and predicting BGLs. The neural network is pretrained to approximate the solutions of the physiological dynamics, and the output of this pretrained model, representing the insulin and glucose concentration states, is then fed as input into a predictive model, enabling simultaneous optimization of predictive accuracy and physiological parameter estimation, offering advantages over traditional modeling approaches in terms of personalized prediction and interpretability. The results highlight the model’s ability to estimate physiological parameters while maintaining strong predictive performance that aligns with the underlying physiological principles. This framework offers significant potential for personalized predictive modeling where precise and efficient understanding of individual metabolism is essential.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040095

Authors: Carlos Martínez Blanca Posada Olivia Zulaica Laura Busto Joaquín Triñanes César Veiga

Understanding Alzheimer’s Disease (AD) progression using resting-state functional Magnetic Resonance Imaging (rs-fMRI) remains an open challenge. Variational Autoencoders (VAEs) provide compact representations of high-dimensional neuroimaging data but lack mechanisms to highlight disease-relevant features. We propose KS-VAE, a novel framework that integrates the Kolmogorov–Smirnov test into the latent space of VAEs to identify statistically significant variables discriminating healthy from pathological brain states. This integration enables measurement of latent space shifts associated with cognitive decline, offering a quantitative approach to neurodegenerative processes. By modifying the most relevant variables, KS-VAE generates synthetic samples that simulate transitions between clinical conditions while preserving anatomical plausibility. The method enhances the modeling of temporal and distributional dynamics underlying disease progression and provides interpretable analysis of class-relevant features. Applied to rs-fMRI scans of 220 subjects from the ADNI cohort, KS-VAE demonstrated robust class separation between cognitively normal and Alzheimer’s disease subjects, achieving a classification accuracy of 84.5% and an F1-score of 84.5%, and clinically consistent synthetic transitions. KS-VAE thus offers a statistically grounded and clinically interpretable framework for understanding Alzheimer’s disease progression.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040094

Authors: Vivek Dwivedi Gregor Rozinaj Javlon Tursunov Ivan Minárik Marek Vanco Radoslav Vargic

Sparse light field imaging often limits the quality of 3D scene reconstruction due to insufficient viewpoint coverage, resulting in incomplete or inaccurate reconstructions. This work introduces a hybrid CNN–LSTM-based framework to address this issue by generating novel camera poses and the corresponding synthesized novel views, effectively densifying the light field representation. The CNN extracts spatial features from the sparse input views, while the LSTM predicts temporal and positional dependencies, enabling smooth interpolation of novel poses and views. The proposed method integrates these synthesized views with the original sparse dataset to produce a comprehensive set of images. Our approach was evaluated on several datasets, including challenging datasets. The inference capability of our method was tested extensively, and it showed good generalization across diverse datasets. The effectiveness of the framework was evaluated not only with local light field fusion (LLFF) but also with NeRF and 3D Gaussian Splatting, which are considered state-of-the-art reconstruction methods. Overall, the enriched dataset generated by our method led to consistent improvements in 3D reconstruction quality, including higher depth estimation accuracy, reduced artifacts, and enhanced structural consistency. Most importantly, LSTM-based approaches have so far attracted limited attention in the context of generating novel views. While LSTMs have been widely applied in sequential data domains such as natural language processing, their use for image generation conditioned on camera poses remains largely unexplored, which underscores the novelty and significance of the proposed work. This approach provides a scalable and generalizable solution to the sparsity problem in light fields, advancing the capabilities of computational imaging, photorealistic rendering, and immersive 3D scene reconstruction. The results firmly establish the proposed method as a robust and versatile tool for improving reconstruction quality in sparse-view settings.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040093

Authors: Flavia Pennisi Antonio Pinto Claudia Cozzolino Andrea Cozza Giovanni Rezza Carlo Signorelli Vincenzo Baldo Vincenza Gianfredi

Background: Early differentiation of mosquito-borne viral infections from other causes of acute febrile illness remains challenging, particularly in endemic and resource-limited settings. Artificial intelligence (AI) models have been proposed to improve early diagnosis, but their incremental value over conventional approaches is unclear. Methods: We conducted a systematic review and meta-analysis of comparative studies evaluating AI/machine learning models versus conventional approaches (clinical assessment, laboratory-based pathways, or traditional statistical models) for early detection of mosquito-borne viral infections. PubMed, Embase, and Scopus were searched through August 2025. Paired performance metrics were synthesized using fixed- and random-effects models. Outcomes included AUC, sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). Risk of bias was assessed using PROBAST. Results: Thirteen studies met inclusion criteria. Under random-effects models, AI improved sensitivity (ES = 2.64, p = 0.028), specificity (ES = 5.55, p < 0.001), accuracy (ES = 3.19, p < 0.001), and NPV (ES = 13.84, p < 0.001). No consistent advantage was observed for AUC, and PPV findings were inconsistent. Substantial heterogeneity was present across outcomes (I2 = 100%). Most studies relied on internal validation, and PROBAST identified high risk of bias in the analysis domain in over half. Conclusions: AI-based models may enhance threshold-dependent performance metrics, supporting their use as adjunctive decision-support tools for early triage and case exclusion, while external validation and implementation-focused research remain essential.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040092

Authors: Pavlo Radiuk Oleksander Barmak Leonid Bedratyuk Iurii Krak

Deep learning systems deployed in regulated settings require explanations that are accurate and stable under nuisance transformations, yet classical post hoc transition matrices rely on fidelity-only fitting that fails to guarantee consistent explanations under spatial rotations or other group actions. In this work, we propose Equivariant Transition Matrices, a post hoc approach that augments transition matrices with Lie-group-aware structural constraints to bridge this research gap. Our method estimates infinitesimal generators in the formal and mental feature spaces, enforces an approximate intertwining relation at the Lie algebra level, and solves the resulting convex Least-Squares problem via singular value decomposition for small networks or implicit operators for large systems. We introduce diagnostics for symmetry validation and an unsupervised strategy for regularization weight selection. On a controlled synthetic benchmark, our approach reduces the symmetry defect from 13,100 to 0.0425 while increasing the mean squared error marginally from 0.00367 to 0.00524. On the MNIST dataset, the symmetry defect decreases by 72.6 percent (141.19 to 38.65) with changes in structural similarity and peak signal-to-noise ratio below 0.03 percent and 0.06 percent, respectively. These results demonstrate that explanation-level equivariance can be reliably imposed post-training, providing geometrically consistent interpretations for fixed deep models.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040091

Authors: Pavel Posokhov Stepan Skrylnikov Sergei Masliukhin Alina Zavgorodniaia Olesia Koroteeva Yuri Matveev

The modern information retrieval field increasingly relies on hybrid search systems combining sparse retrieval with dense neural models. However, most existing hybrid frameworks employ static mixing coefficients and independent component training, failing to account for the specific needs of individual queries and corpus heterogeneity. In this paper, we introduce an adaptive hybrid retrieval framework featuring query-driven alpha prediction that dynamically calibrates the mixing weights based on query latent representations instantiated in a lightweight low-latency configuration and a full-capacity encoder-scale predictor, enabling flexible trade-offs between computational efficiency and retrieval accuracy without relying on resource-inefficient LLM-based online evaluation. Furthermore, we propose antagonist negative sampling, a novel training paradigm that optimizes the dense encoder to resolve the systematic failures of the lexical retriever, prioritizing hard negatives where BM25 exhibits high uncertainty. Empirical evaluations on large-scale multilingual benchmarks (MLDR and MIRACL) indicate that our approach demonstrates superior average performance compared to state-of-the-art models such as BGE-M3 and mGTE, achieving an nDCG@10 of 74.3 on long-document retrieval. Notably, our framework recovers up to 92.5% of the theoretical oracle performance and yields significant improvements in nDCG@10 across 16 languages, particularly in challenging long-context scenarios.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040090

Authors: Vukašin Ćirović Vesna Ranković Nikola Milivojević Vladimir Milivojević Brankica Majkić-Dursun

Dam monitoring data are nonlinear and nonstationary time series. Most existing data-driven dam displacement models are developed independently for each measuring point, disregarding the fact that a dam is a complex structure composed of various interconnected elements that form a unified whole. Regardless of the dam type, all points on the dam are exposed to the same external environmental influences. To account for the correlation between displacement time series at different points, this paper proposes a novel fine-tuned deep-learning nonlinear autoregressive (NAR) model based on a Long Short-Term Memory (LSTM) network for predicting dam tangential displacement, and a new method for generating source data to train the base model. The models for three measuring points were developed and tested on experimental data collected over a period of slightly more than twelve years. Compared with the model without fine-tuning, the proposed approach achieves an average mean square error (MSE) reduction of 80.68% on the training set and 65.79% on the test set, as well as an average mean absolute error (MAE) reduction of 51.05% and 52.62%, respectively. Furthermore, the proposed model outperforms Random Forest (RF), Support Vector Regression (SVR), and Multi-Layer Perceptron (MLP) models for dam displacement prediction.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040089

Authors: Timothy Joseph Ahmed Dhaouadi Jayroop Ramesh Assim Sagahyroon Fadi Aloul

Diabetes mellitus is a chronic condition that frequently leads to severe complications that are difficult to detect in their early stages using conventional clinical monitoring. This paper presents a data-driven framework for predicting multiple diabetes-related complications using structured electronic health record data while ensuring clinically meaningful explainability. The proposed approach adapts a pretrained electronic health record foundation model to operate on static patient data and integrates it with classical machine learning baselines to address class imbalance, feature sparsity, and interpretability challenges. A multi-label prediction setting covering eight common diabetes complications is evaluated using a real-world dataset from a regional diabetes center in the United Arab Emirates. Synthetic data generation and clinical constraint enforcement are applied to improve robustness for underrepresented outcomes, while feature selection is guided by model importance and attribution-based explanations. The best-performing configuration, a weighted ensemble combining a low-rank adapted Hyena-based foundation model with a tree-based predictor, achieved an average F1-score of 0.77, an average recall of 0.85, and an example-based F1-score of 0.71, outperforming all individual models. In addition, this ensemble produced the most stable explanations under input perturbations, indicating improved consistency of dominant clinical risk drivers. These results demonstrate that explainable foundation model-based ensembles can deliver accurate, robust, and clinically transparent risk prediction for diabetes complications.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040088

Authors: Shaheen Khatoon Jibran Mir Azhar Mahmood

Named Entity Recognition (NER) plays a crucial role in Aspect-Based Sentiment Identification (ABSI), enabling the extraction of domain-specific aspects and their associated sentiment expressions from unstructured textual reviews. In complex domains such as movie reviews, sentiment is frequently conveyed through references to named entities (e.g., actors, directors, or movie titles) and other contextual cues. However, many existing ABSI approaches treat NER as a separate preprocessing step, limiting the effective modeling of entity–aspect–opinion relationships. Integrating NER directly into the ABSI framework, allows entity-specific opinions to be more accurately identified, overlapping aspects to be disambiguated, and contextual sentiment expressions to be captured more effectively. To address these challenges, this study proposes an integrated NER-based aspect identification model built on feature-enhanced LSTM and BiLSTM architectures. Linguistic features, including Parts-of-Speech (POS) tags and chunking information, are incorporated to enrich contextual representations, while a Conditional Random Field (CRF) decoding layer models inter-label dependencies for coherent sequence-level predictions of named entities, aspects, and associated opinion expressions. Compared with large transformer-based models, the proposed BiLSTM-CRF architecture offers lower computational complexity, fewer parameters, and allows explicit integration and analysis of linguistic features that are often implicitly encoded in transformer attention mechanisms. The model is evaluated through multiple experimental variants across three domains. Four configurations are applied to movie-review data to jointly extract person names, movie titles, and aspect-opinion pairs, while six configurations assess cross-domain robustness on restaurant and laptop review datasets. Results show that the BiLSTM-CRF model augmented with POS features consistently outperforms baseline configurations in the movie domain and remains competitive across domains, achieving an F1-score of 0.89. These findings demonstrate that explicit linguistic feature integration within a CRF-based sequence modeling can provide an effective and computationally efficient alternative to large-scale transformer fine-tuning for structured, entity-linked ABSI tasks.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040087

Authors: Mohammad Fazle Rabbi

Post-combustion amine scrubbing using monoethanolamine (MEA) remains a leading carbon capture technology, yet its deployment is constrained by high regeneration energy requirements and the computational expense of rigorous process simulation. This study presents an integrated framework coupling high-fidelity rate-based process simulation with explainable machine learning to systematically characterize a ten-dimensional operating space for MEA-based CO2 absorption. Latin hypercube sampling generated 10,000 steady-state cases, and five regression architectures were benchmarked under identical protocols. A neural network achieved the highest accuracy (R2 = 0.9729, RMSE = 1.43%), while XGBoost was selected as the operational surrogate due to its robust computational efficiency (1.5 ms inference latency) and native compatibility with exact Shapley value decomposition. SHAP analysis identified liquid-to-gas ratio as the dominant efficiency determinant, contributing 46.6% of total predictive importance, followed by inlet temperature and MEA concentration, with these three parameters collectively explaining 85% of efficiency variation and establishing a compact control hierarchy suitable for reduced-order control architectures. Bivariate interaction analysis located a high-efficiency operating region, while sensitivity analysis confirmed the strong influence of inlet temperature across the operating envelope. Pareto optimization via NSGA-II generated tiered operational guidelines spanning the 85% to 98% capture efficiency range, quantifying a 39% specific regeneration duty penalty (3.1 to 4.3 MJ/kg CO2) for pursuing maximum versus baseline capture targets. The framework demonstrates how explainable machine learning converts opaque process simulations into actionable engineering knowledge, providing a transparent and computationally efficient basis for design optimization and digital twin deployment in post-combustion carbon capture systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040086

Authors: Yanche Ari Kustiawan Khairil Imran Ghauth

Phishing remains a major cybersecurity threat, yet the application of quantum machine learning (QML) to phishing detection is still at an early stage. This study presents a systematic literature review aimed at providing a concise overview of existing QML-based approaches for phishing detection, identifying methodological trends, limitations, and future research directions. A PRISMA-guided review protocol was applied to peer-reviewed journal and conference articles published between 2021 and 2025, retrieved from major scientific databases. Eligible studies were analyzed in terms of QML models, feature encoding strategies, experimental settings, evaluation metrics, and study quality using an adapted Newcastle–Ottawa Scale. The results indicate that current research is limited in volume and largely focuses on hybrid quantum–classical models, particularly quantum support vector machines and variational quantum classifiers. Reported performance is highly dependent on encoding methods, circuit depth, and simulator-based experimentation, with few studies evaluating real quantum hardware. Common challenges include small datasets, lack of external validation, hardware noise, scalability constraints, and the absence of standardized benchmarks. Overall, the review suggests that QML for phishing detection remains exploratory and is not yet competitive with mature classical approaches, but it holds potential as an experimental research direction, provided that future studies address robustness, reproducibility, and practical deployment constraints.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040085

Authors: Kevin Mallinger Sebastian Raubitzek Sebastian Schrittwieser Edgar Weippl

Noise contamination is a common challenge in the analysis of time series data, where stochastic perturbations can obscure deterministic dynamics and complicate the interpretation of signals from chaotic and physiological systems. Reliable identification of noise regimes and their intensity is therefore essential for robust analysis of dynamical and biomedical signals, where incorrect attribution of stochastic perturbations can lead to misleading interpretations of system behavior. For this reason, the present study examines the role of complexity-based descriptors for identifying stochastic perturbations in time series and analyzes how these metrics respond to different noise regimes across heterogeneous dynamical systems. A supervised learning approach based on complexity descriptors was developed to analyze controlled perturbations in multiple signal types. Gaussian, pink, and low-frequency noise disturbances were injected at predefined intensity levels into the Rössler and Lorenz chaotic systems, the Hénon map, and synthetic electrocardiogram signals, while AR(1) processes were used for validation on inherently stochastic signals. From these systems, eighteen entropy-based, fractal, statistical, and singular value decomposition-based complexity metrics were extracted from either raw signals or reconstructed phase spaces. These features were used to perform three classification tasks that capture different aspects of noise characterization, including detecting the presence of noise, identifying the perturbation type, and discriminating between different noise intensities. In addition to predictive modeling, the study evaluates the complexity profiles and feature relevance of the metrics under varying perturbation regimes. The results show that no single complexity metric consistently discriminates noise regimes across all systems. Instead, system-specific relevance patterns emerge. Under given experimental constraints (data partitioning, machine learning algorithm, etc.), Approximate Entropy provides the strongest discrimination for the Lorenz system and the Hénon map, the Coefficient of Variation, Sample and Permutation Entropy dominate classification for ECG signals, and the Condition Number and Variance of first derivative together with Fisher Information are most informative for the Rössler system. Across all datasets, the proposed framework achieves an average accuracy of 99% for noise presence detection, 98.4% for noise type classification, and 98.5% for noise intensity classification. These findings demonstrate that complexity metrics capture structural and statistical signatures of stochastic perturbations across a diverse set of dynamic systems.

Machine Learning and Knowledge Extraction doi: 10.3390/make8040084

Authors: Corneliu Florea Eduard Barnoviciu

In this paper, we present an extensive evaluation of tabular data distillation methods for downstream classification and regression tasks. Our analysis considers multiple distillation approaches that are problem-type independent (i.e., unsupervised). For downstream learners, we focus on non-neural models such as Random Forest, XGBoost, and Support Vector Machines, as our goal is to evaluate the quality of the distilled data independently of the learner. The evaluation is conducted on 17 classification and nine regression problems. Our findings can be summarized as follows: (1) in all cases, applying a distillation method leads to a decrease in performance compared to the baseline; (2) overall, coreset-based methods are the most effective, with performance losses that are minimal—ranging from around 3% in classification accuracy or regression correlation to, in some cases, being negligible; (3) performance loss is moderately correlated with dataset tailness, measured as the proportion of outliers; (4) all distillation methods alter dataset consistency, narrowing the range of hyperparameter values that yield good performance; and (5) the Coreset Leverage Score remains fast, regardless of the size of the original set and of the distilled set.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030083

Authors: Wenshuai Wang Hong Liu Shengquan Li Peifeng Jiang Dandan Che Runwei Ding

Visual localization plays a critical role for mobile robots to estimate their position and orientation in GPS-denied environments. However, its efficiency, robustness, and generalization are fundamentally undermined by severe viewpoint changes and dramatic appearance variations, which present persistent challenges for image-based feature representation and pose estimation under real-world conditions. Recently, map-free visual relocalization (MFVR) has emerged as a promising paradigm for lightweight deployment and privacy isolation on edge devices, while how to learn compact and invariant image tokens without relying on structural 3D maps still remains a core problem, particularly in highly dynamic or long-term scenarios. In this paper, we propose the Debiased Multiplex Tokenizer as a novel method (termed as DMT-Loc) for efficient and versatile MFVR to address these issues. Specifically, DMT-Loc is built upon a pretrained vision Mamba encoder and integrates three key modules for relative pose regression: First, Multiplex Interactive Tokenization yields robust image tokens with non-local affinities and cross-domain descriptions. Second, Debiased Anchor Registration facilitates anchor token matching through proximity graph retrieval and autoregressive pointer attribution. Third, Geometry-Informed Pose Regression empowers multi-layer perceptrons with a symmetric swap gating mechanism operating inside each decoupled regression head to support accurate and flexible pose prediction in both pair-wise and multi-view modes. Extensive evaluations across seven public datasets demonstrate that DMT-Loc substantially outperforms existing baselines and ablation variants in diverse indoor and outdoor environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030082

Authors: Suliman Thwib Radwan Qasrawi Ghada Issa Razan AbuGhoush Hussein AlMasri Marah Qawasmi

Background: Real-time and accurate analysis of breast ultrasounds is crucial for diagnosis but remains challenging due to issues like low image contrast and operator dependency. This study aims to address these challenges by developing an integrated framework for real-time lesion detection and tracking. Methods: The proposed system combines Contrast-Limited Adaptive Histogram Equalization (CLAHE) for image preprocessing, a transfer learning-enhanced YOLOv11 model following a continual learning paradigm for cross-center generalization in for lesion detection, and a novel Detection-Based Tracking (DBT) approach that integrates Kernelized Correlation Filters (KCF) with periodic detection verification. The framework was evaluated on a dataset comprising 11,383 static images and 40 ultrasound video sequences, with a subset verified through biopsy and the remainder annotated by two radiologists based on radiological reports. Results: The proposed framework demonstrated high performance across all components. The transfer learning strategy (TL12) significantly improved detection outcomes, achieving a mean Average Precision (mAP) of 0.955, a sensitivity of 0.938, and an F1 score of 0.956. The DBT method (KCF + YOLO) achieved high tracking accuracy, with a success rate of 0.984, an Intersection over Union (IoU) of 0.85, and real-time operation at 54 frames per second (FPS) with a latency of 7.74 ms. The use of CLAHE preprocessing was shown to be a critical factor in improving both detection and tracking stability across diverse imaging conditions. Conclusions: This research presents a robust, fully integrated framework that bridges the gap between speed and accuracy in breast ultrasound analysis. The system’s high performance and real-time efficiency underscore its strong potential for clinical adoption to enhance diagnostic workflows, reduce operator variability, and improve breast cancer assessment.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030081

Authors: Moritz Moß Sebastian Boldt Gurbandurdy Dovletov Adjie Salman Josef Pauli Dietmar Lerche Marco Gleiß Hermann Nirschl Johannes Walter Wolfgang Peukert

Applications for machine learning (ML) and deep learning (DL) are constantly growing and have already been adopted in the field of particle measurement technology. Even though analytical (ultra-)centrifugation (AC/AUC) is a widely used technique for characterizing dispersed particle systems, ML and DL have not yet been applied in this area. Data evaluation and interpretation in AC/AUC can be challenging and often requires expert knowledge. DL models can help, but their development is limited by a lack of annotated training data. One solution is to generate and use synthetic data instead. In the first part of this study, a model was trained to synthesize data from experiments using a combination of Variational Autoencoder (VAE) and Generative Adversarial Networks (GANs). The results appear highly realistic. Novice users could distinguish real from synthetic samples with only 63% accuracy. Then, a classifier was trained on experimental AC data to categorize real-world examples based on their underlying separation kinetics, testing different DL architectures. After initial training, the models were further fine-tuned with synthetic AC data. ResNet34 models achieved the best performance with 94% accuracy, comparable to an AC expert (91%), while inexperienced users reached only 53%. In the second part of our study, a regression model was trained for the analysis of sedimentation coefficients. Therefore, various generative models were developed and evaluated for synthesizing AUC data based on numerically simulated sedimentation boundaries. The best results were achieved by combining VAE and GAN architectures with embedded physical constraints. However, the generative networks have so far led to additional smearing of the profiles, resulting in a broadening of the sedimentation coefficient distribution and indicating that further refinement is necessary.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030080

Authors: Arley Magnolia Aquino-García Humberto Pérez-Espinosa Javier Andreu-Perez Ansel Y. Rodríguez González

The development of AI-driven computational methods has enabled rapid and non-invasive analysis of respiratory sounds using acoustic data, particularly cough recordings. Although the COVID-19 pandemic accelerated research on cough-based acoustic analysis, many early studies were limited by insufficient data quality, lack of standardized protocols, and limited reproducibility due to data scarcity. In this study, we propose an audio analysis framework for cough-based respiratory disease screening research using COVID-19 as a clinically validated case dataset. All analyses were conducted on a single clinically acquired multicentric dataset collected under standardized conditions in certified laboratories in Mexico and Spain, comprising cough recordings from 1105 individuals. Model training and testing were performed exclusively within this dataset. The framework incorporates signal preprocessing and a comparative evaluation of segmentation strategies, showing that segmented cough analysis significantly outperforms full-signal analysis. Class imbalance was addressed using the Synthetic Minority Over-sampling Technique (SMOTE) for CNN2D models and the supervised Resample filter implemented in WEKA for classical machine learning models, both applied exclusively to the training subset to generate balanced training sets and prevent data leakage. Feature extraction and classification were carried out using Random Forest, Support Vector Machine (SVM), XGBoost, and a 2D Convolutional Neural Network (CNN2D), with hyperparameter optimization via AutoML. The proposed framework achieved a best balanced screening performance of 85.58% sensitivity and 86.65% specificity (Random Forest with GeMAPSvB01), while the highest-specificity configuration reached 93.90% specificity with 18.14% sensitivity (CNN2D with SMOTE and AutoML). These results demonstrate the methodological feasibility of the proposed framework under the evaluated conditions.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030079

Authors: Muhammad Abulaish Anjali Bhardwaj

Emotion classification is a fundamental task in affective computing, with applications in human–computer interaction, mental health monitoring, and social media analysis. Although most existing methods formulate it as a flat classification problem, emotional expressions are inherently structured and grounded in semantic roles such as the emotion cue, stimulus, experiencer, and target. However, the relative contribution of these roles to emotion inference has not been systematically examined. Unlike prior models, we propose RolEmo, a role-aware framework for emotion classification that explicitly incorporates semantic role information. The framework employs a controlled role-masking strategy to analyze the contribution of individual roles, augments textual representations with external commonsense knowledge to capture implicit affective context, and applies supervised contrastive learning to structure the embedding space by bringing emotionally similar instances closer while separating opposing ones. We evaluate RolEmo on three benchmark datasets annotated with semantic roles. Experimental results demonstrate that RolEmo outperforms the strongest baseline across three datasets by up to 16.4%, 25.8%, and 23.2% in the Full Text, Only Role, and Without Role settings, respectively. The analysis further indicates that the cue and stimulus roles provide the most reliable signals for emotion classification, with their removal causing performance drops of up to 6.2% in macro f1-score, while experiencer and target roles exhibit more variable effects. These findings highlight the importance of structured semantic modeling and commonsense reasoning for robust and interpretable emotion understanding.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030078

Authors: Dominik Polke Tim Kösters Elmar Ahle Dirk Söffker

In complex and unknown processes, global models are fitted over the entire input domain but often tend to perform poorly whenever the response surface exhibits non-stationary behavior and varying smoothness. A common approach is to use local models, which requires partitioning the input domain into subdomains and training multiple models, thereby adding significant complexity. Recognizing this limitation, this study addresses the need for models that represent the input–output relationship consistently over the full domain while still adapting to local variations in the response. It introduces a novel machine learning approach: the Polynomial Chaos Expanded Gaussian Process (PCEGP), leveraging polynomial chaos expansion to calculate input-dependent hyperparameters of the Gaussian process (GP). This provides a mathematically interpretable approach that incorporates non-stationary covariance functions and heteroscedastic noise estimation to generate locally adapted models. The model performance is compared to different algorithms in benchmark tests for regression tasks. The results demonstrate low prediction errors of the PCEGP, highlighting model performance that is often competitive with or better than previous methods. A key advantage of the presented model is its interpretable hyperparameters along with training and prediction runtimes comparable to those of a standard GP.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030077

Authors: Zhe-Yu Ye Jun-Mu Peng Bing-Qian Lu Tamotsu Kamishima

Manual single-slice lumbar quantitative computed tomography (QCT) depends on operator-driven slice selection and trabecular region-of-interest (ROI) placement. We developed a fully automated single-slice workflow for vertebral trabecular Hounsfield unit (HU) measurement that combines unsuitable-slice prescreening, dual-purpose segmentation, intra-patient slice-quality ranking, and a deterministic inner ROI rule. The pipeline includes an Eligibility Gate, QC-Envelope segmentation for broad, vertebral- and usability-preserving delineation, PairRank-Swin for best-slice selection, and dedicated trabecular segmentation for final quantitative analysis. In the independent external cohort, 4 cases were considered non-evaluable by both manual review and the pipeline, and 2 additional borderline-quality cases were manually measured but rejected by the pipeline; therefore, paired HU agreement analysis included 44 evaluable cases. Agreement remained high, with Pearson’s r = 0.987, Lin’s CCC = 0.985, mean bias −0.44 HU, and limits of agreement from −14.88 to +13.99 HU. Coverage was 84.1% within ±10 HU and 97.7% within ±15 HU. Ablation analysis showed that slice ranking and ROI erosion were the most critical components. In an open module-level baseline comparison, QC-Envelope segmentation substantially outperformed TotalSegmentator. This workflow provides high agreement with expert HU measurement while preserving reviewable intermediate outputs.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030076

Authors: Anthony McCofie Dmitry Goldgof Jacqueline Hausmann Peter R. Mouton Yu Sun Md Imran Hossain

Infant cries serve as a primary indicator of distress and pain; however, distinguishing pain-related cries from those triggered by other needs remains a challenging task, even for trained professionals. Timely and accurate pain assessment is essential for appropriate medical intervention, particularly in preverbal infants who cannot express their needs verbally. Recently, Deep Learning (DL) models have demonstrated significant potential in addressing this challenge by enabling automated and efficient pain assessment through audio signal processing. In this survey, we review methods for pain assessment from infant cry sounds, covering deep learning architectures, modern Transformer-based models, and emerging Vision-Language Model (VLM) pipelines. The review includes approaches that integrate Mel-spectrogram representations of cry audio with multimodal model frameworks to improve robustness, interpretability, and cross-modal reasoning in pain detection. By summarizing recent advancements and identifying limitations and open challenges in current methodologies, this review aims to provide insights into future research directions that may enhance the robustness, generalizability, and clinical applicability of automated infant pain assessment tools.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030075

Authors: Lili Zhang Quanyan Zhu

We propose a centralized Activity Retrieval and Detection Intelligence (ActivityRDI) solution framework, demonstrate its application performance in network threat detection in detail, and show its generalization to other domains. Network threat detection is challenging owing to the complex nature of attack activities and the limited historically revealed threat data from which to learn. To help enhance the existing methods (e.g., analytics, machine learning, and artificial intelligence) to detect the network threats, we propose a multi-agent AI solution for agile threat detection. In this solution, a knowledge graph is used to analyze changes in user activity patterns and calculate the risk of unknown threats. Then, an imbalanced learning model is used to prune and weight the knowledge graph and to calculate the risk of known threats. Finally, a large language model (LLM) is used to retrieve and interpret the risk associated with user activities from the knowledge graph and the imbalanced learning model. The preliminary results show that the solution improves the threat capture rate by 3–4% and adds natural language interpretations of the risk predictions based on user activities with 95% accuracy. Furthermore, a demonstration application has been built to show how the proposed solution framework can be deployed and used. The generalizability of the proposed solution in other domains is also shown through an application to customer engagement, with 97% accuracy.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030074

Authors: Luke Korthals Emma Akrong Gali Geller Hannes Rosenbusch Raoul Grasman Ingmar Visser

Large language models (LLMs) show promise for grading open-ended assessments but still exhibit inconsistent accuracy, systematic biases, and limited reliability across assignments. To address these concerns, we introduce SURE (Selective Uncertainty-based Re-Evaluation), a human-in-the-loop pipeline that combines repeated LLM prompting, uncertainty-based flagging, and selective human regrading. Three LLMs—gpt-4.1-nano, gpt-5-nano, and the open-source gpt-oss-20b—graded answers of 46 students to 130 open questions and coding exercises across five assignments. Each student answer was scored 20 times to derive majority-voted predictions and self-consistency-based certainty estimates. We simulated human regrading by flagging low-certainty cases and replacing them with scores from four human graders. We used the first assignment as a training set for tuning certainty thresholds and to explore LLM output diversification via sampling parameters, rubric shuffling, varied personas, multilingual prompts, and post hoc ensembles. We then evaluated the effectiveness and efficiency of SURE on the other four assignments using a fixed certainty threshold. Across assignments, fully automated grading with a single prompt resulted in substantial underscoring, and majority-voting based on 20 prompts improved but did not eliminate this bias. Low certainty (i.e., high output diversity) was diagnostic of incorrect LLM scores, enabling targeted human regrading that improved grading accuracy while reducing manual grading time by 40–90%. Aggregating responses from all three LLMs in an ensemble improved certainty-based flagging and most consistently approached human-level accuracy, with 70–90% of the grades students would receive falling inside human-grader ranges. A reanalysis based on outputs from a more diversified LLM ensemble comprised of gpt-5, codestral-25.01, and llama-3.3-70b-instruct replicated these findings but also suggested that large reasoning models such as gpt-5 might eliminate the need for human oversight of LLM grading entirely. These findings demonstrate that self-consistency-based uncertainty estimation and selective human oversight can substantially improve the reliability and efficiency of AI-assisted grading.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030073

Authors: Michel Planat

Recent work on the evaluation of large language models emphasizes that the relevant unit of intelligence is not the artificial system alone but the human–AI hybrid. In parallel, topological and dynamical models of cognition based on Painlevé equations and non-semisimple topology propose that consciousness, intelligence, and creativity emerge from constrained long-horizon dynamics near criticality. This perspective article argues that these two research directions are deeply compatible. We show that the empirical framework for human–AI collaboration can be interpreted as a fusion process between complementary cognitive sectors: exploration (AI) and selection (human cognition). The dynamical mechanism underlying this fusion is identified with noisy phase locking between cognitive oscillators. Two independent routes to a universal 1/f spectral signature are developed: a geometric route through the WKB/Stokes analysis of Painlevé V confluence, and an arithmetic route through the Mangoldt function and harmonic interactions in phase-locked loops. We connect these results to the Bost–Connes quantum statistical model, whose phase transition at the pole of the Riemann zeta function provides an exact mathematical framework for the lock-in phase hypothesis of identity consolidation in AI systems. This synthesis suggests a unified research program for hybrid intelligence grounded in topology, dynamical systems, number theory, and real-world AI evaluation.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030072

Authors: Xeniya Bashkova Anastasia Molodtsova Nikita Olekhno Alexander Hvatov

Active matter actively searches for models that allow them to connect the behavior of multiple agents to particle system with a physical law. However, the arsenal of models used to model active matter systems is very restricted. Modern differential equation discovery approaches allow one to extract governing equations from data for a single particle in the form of the ODE. However, there is still the question of how to model at the meso- and macroscales. This paper presents a data-driven framework for extracting the governing physical laws of a hardware-made swarm across multiple scales of organization. Using the EPDE framework, we transition from a discrete, chaotic trajectory of individual agents to a continuous, effective field theory of the collective. We show that augmenting the symbolic search space with interaction-aware tokens allowed for the derivation of stochastic partial differential equations (SDEs) that significantly outperformed baseline deterministic models (reducing CRPS by up to 10%). Additionally, we derive a system of SPDEs that governs the macroscale displacement field.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030071

Authors: Ensieh Alipour Seyed Mohammad-Bagher Malaek

This study presents a novel data-driven framework for developing airport-specific landing policies and procedures from historical successful-landing data. The proposed process, termed the Airport-Dependent Landing Procedure (ADLP), is motivated by the fact that airports rely on uniquely tailored approach charts reflecting local operational constraints and environmental conditions. While existing approach charts and landing procedures are primarily designed based on expert knowledge, safety margins, and regulatory conventions, the authors argue that data science and data mining techniques offer a complementary and empirically grounded methodology for extracting operationally meaningful structures directly from historical landing data. In this work, we construct a probabilistic three-dimensional environment from real-world aircraft approach trajectories, capturing spatiotemporal relationships under varying atmospheric conditions during approach. The proposed methodology integrates Adversarial Inverse Reinforcement Learning (AIRL) with Recurrent Proximal Policy Optimization (R-PPO) to establish a foundation for automated landing without pilot intervention. AIRL infers reward functions that are consistent with behaviors exhibited in prior successful landings. Subsequently, R-PPO is employed to learn control policies that satisfy safety constraints related to airspeed, sink rate, and runway alignment. Application of the proposed framework to real approach trajectories at Guam International Airport demonstrates the efficiency and effectiveness of the methodology.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030070

Authors: Ponciano Jorge Escamilla-Ambrosio Brandon Iván Méndez-Barrera Alberto Jorge Rosales-Silva Gina Gallegos-García Gilberto Lorenzo Martínez-Luna

The increasing complexity and heterogeneity of Internet of Things (IoT) systems pose significant challenges for systematic security and vulnerability assessment. From a knowledge-centric perspective, IoT security analysis requires transforming heterogeneous asset information into structured and interpretable security knowledge. In this paper, we propose a structured methodology for vulnerability analysis that models the attack surface of an IoT system by explicitly linking asset characteristics to known vulnerabilities, security controls, and countermeasures. The approach starts with a visual representation of the system architecture, where hardware, software, and communication components are identified and described through their technical characteristics. These characteristics are automatically mapped to relevant vulnerabilities, security controls, and countermeasures using a dedicated software tool called AVCA (Asset Vulnerabilities and Countermeasures Analyzer). The tool generates graph-based analytical representations that model vulnerabilities–countermeasures relationships in compliance with the Cloud Security Alliance (CSA) IoT Security Framework. From these graphs, attack–countermeasure trees are derived to provide a clear and interpretable representation of potential threats and mitigation strategies. The proposed methodology was evaluated through a case study involving a representative IoT system and an exploratory applicability experiment with participants with different levels of experience in IoT and cybersecurity. The results suggest that the approach is feasible and practically applicable for supporting security analysts in the systematic assessment of IoT attack surfaces, vulnerability identification, and selection of appropriate countermeasures under the evaluated conditions. This work highlights the role of structured and interpretable knowledge extraction as a foundation for knowledge-centric and interpretable IoT security analysis.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030069

Authors: Francisco Herrera Salvador García María José del Jesus Luciano Sánchez Marcos López de Prado

Human–AI collaboration (HAIC) increasingly mediates high-risk decisions in public and private sectors, yet many documented AI harms arise not only from model error but from breakdowns in joint human–AI work: miscalibrated reliance, impaired contestability, misallocated agency, and governance opacity. Conventional explainable AI (XAI) approaches, often delivered as static one-shot artifacts, are poorly matched to these sociotechnical dynamics. This paper is a position paper arguing that explainability should be reframed as a harm-mitigation infrastructure for HAIC: an interactive, iterative capability that supports ongoing sensemaking, safe handoffs of control, governance stakeholder roles and institutional accountability. We introduce co-explainers as a conceptual framework for interactive XAI, in which explanations are co-produced through structured dialogue, feedback, and governance-aware escalation (explain → feedback → update → govern). To ground this position, we synthesize prior harm taxonomies into six HAIC-oriented harm clusters and use them as heuristic design lenses to derive cluster-specific explainability requirements, including uncertainty communication, provenance and logging, contrastive “why/why-not” and counterfactual querying, role-sensitive justification, and recourse-oriented interaction protocols. We emphasize that co-explainers do not “mitigate” sociotechnical harms in isolation; rather, they provide an interface layer that makes harms more detectable, decisions more contestable, and accountability handoffs more operational under realistic constraints such as sealed models, dynamic updates, and value pluralism. We conclude with an agenda for evaluating co-explainers and aligning interactive XAI with governance frameworks in real-world HAIC deployments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030068

Authors: Abdalla Y. Almarzooqi Mohamed G. Arab Maher Omar Emran Alotaibi

Dynamic characterization of calcareous (carbonate) sands is essential for performance-based design of offshore foundations, coastal reclamation, and marine infrastructure in tropical and subtropical regions. In contrast to silica sands, carbonate sediments are biogenic and typically comprise angular, irregular grains with intra-particle voids and fragile skeletal microstructure. These traits promote grain crushing and fabric evolution at relatively low-to-moderate confinement, leading to pronounced stress dependency, strong nonlinearity with strain amplitude, and substantial scatter in laboratory stiffness and damping measurements. Consequently, empirical correlations calibrated primarily on quartz sands may yield biased estimates when transferred to carbonate environments. This study presents an ML-driven, leakage-aware benchmarking framework for predicting two key dynamic parameters of biogenic calcareous sands, damping ratio D and shear modulus G, using standard tabular descriptors commonly available in geotechnical practice. Two consolidated experimental databases were curated from resonant column and cyclic triaxial measurements (D: n=890; G: n=966), spanning mean effective confining stress 25 ≤ σm′≤1600 kPa and a wide range of density and gradation conditions. To emphasize transferability, explicit deposit/site labels were excluded, and missingness arising from heterogeneous reporting was handled through a consistent preprocessing pipeline (training-only imputation, categorical encoding, and scaling). Eleven regression algorithms were evaluated, covering linear baselines, regularized regression, neighborhood learning, single trees, bagging and boosting ensembles, kernel regression, and a feedforward neural network. Performance was assessed using R2, RMSE, and MAE on training/validation/test splits, and engineering credibility was supported through explainability-based diagnostics to verify mechanically plausible sensitivities. Results show that ensemble-tree models (Extra Trees and Random Forest) provide the most reliable accuracy–robustness balance across both targets, consistently outperforming linear models and the tested SVR configuration and exhibiting stable validation-to-test behavior. The explainability audit confirms physically meaningful separation of governing controls: stiffness is primarily stress-controlled (σm′ dominant for G), whereas damping is primarily strain-controlled (γ dominant for D). The proposed framework supports practical deployment as a fast surrogate for generating G − γ and D − γ curves within the training domain and for guiding targeted laboratory test planning in carbonate settings.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030067

Authors: Fazliddin Makhmudov Gayrat Juraev Ozod Yusupov Parvina Nasriddinova Dusmurod Kilichev

The rapid growth of Internet of Medical Things (IoMT) devices has revolutionized diagnostics and patient care within smart healthcare networks. However, this progress has also expanded the attack surface due to the heterogeneity and interconnectivity of medical devices. To overcome the limitations of traditional batch-trained security models, this study proposes an adaptive online intrusion detection framework designed for real-time operation in dynamic healthcare environments. The system combines Leveraging Bagging with Hoeffding Tree classifiers for incremental learning while integrating the Page–Hinkley test to detect and adapt to concept drift in evolving attack patterns. A modular and scalable network architecture supports centralized monitoring and ensures seamless interoperability across various IoMT protocols. Implemented within a low-latency, high-throughput stream-processing pipeline, the framework meets the stringent clinical requirements for responsiveness and reliability. To simulate streaming conditions, we evaluated the model using the CICIoMT2024 dataset, presenting one instance at a time in random order to reflect dynamic, real-time traffic in IoMT networks. Experimental results demonstrate exceptional performance, achieving accuracies of 0.9963 for binary classification, 0.9949 for six-class detection, and 0.9860 for nineteen-class categorization. These results underscore the framework’s practical efficacy in protecting modern healthcare infrastructures from evolving cyber threats.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030066

Authors: Shafiullah Qureshi Ba M. Chu Fanny S. Demers Najib Khan Ateeq ur Rehman Irshad

We forecast monthly Canadian real GDP growth using machine learning models trained on Official macroeconomic indicators and Google Trends (GT) data. Predictors are selected dynamically in each rolling window using PDC-SIS, with cross-validation-based tuning to support real-time forecasting and avoid data leakage. The evaluation is conducted on the latest-available (final-vintage) series and should be interpreted as a pseudo out-of-sample forecasting exercise rather than real-time vintage nowcasting. We evaluate GBM, XGBoost, LightGBM, CatBoost, and Random Forest against an ARIMA baseline. Official data deliver the strongest performance at short and medium horizons, while combining Official and GT data yields the clearest improvement at the longest horizon. With GT data alone, LightGBM is the only ML model maintaining positive out-of-sample R2 across all horizons. Diebold–Mariano tests corroborate these patterns: LightGBM dominates other ML models under GT-only predictors, whereas with Official and combined data, the horizon-specific best models significantly outperform ARIMA, with smaller differences among leading tree-based methods.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030065

Authors: Juuso Eronen Michal Ptaszynski Tomasz Wicherkiewicz Robert Borges Katarzyna Janic Zhenzhen Liu Tanjim Mahmud Fumito Masui

Selecting a source language for zero-shot cross-lingual transfer is typically done by intuition or by defaulting to English, despite large performance differences across language pairs. We study whether linguistic similarity can predict transfer performance and support principled source-language selection. We introduce quantified WALS (qWALS), a typology-based similarity metric derived from features in the World Atlas of Language Structures, and evaluate it against existing similarity baselines. Validation uses three complementary signals: computational similarity scores, zero-shot transfer performance of multilingual transformers (mBERT and XLM-R) on four NLP tasks (dependency parsing, named entity recognition, sentiment analysis, and abusive language identification) across eight languages, and an expert-linguist similarity survey. Across tasks and models, higher linguistic similarity is associated with better transfer, and the survey provides independent support for the computational metrics.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030064

Authors: Richard Coric Ebenezer F. Oloyede Heriberto Cuayáhuitl

Medical question answering (QA) systems and conversational agents have attracted growing interest as tools that can assist clinicians, support medical students, and help patients navigate complex information sources. However, existing evaluations of retrieval strategies largely overlook the cost–benefit balance—here referred to as frugality, under realistic computational constraints. This work introduces a frugality-based evaluation framework that jointly assesses accuracy improvements and computational cost to determine when retrieval-augmented generation is beneficial in medical question answering, rather than evaluating retrieval effectiveness through accuracy alone. This study addresses these gaps through a systematic comparative framework that evaluates retrieval relevance, computational efficiency, and knowledge base composition across multiple biomedical QA tasks. We employ open-source LLMs (LlaMA-3-8B-Instruct, Mistral-7B-Instruct-v0.3, and DeepSeek-7B-Chat) across three benchmark medical QA datasets, including MedMCQA, MedQA-USMLE, and PubMedQA. In addition to that, we evaluate a dataset with larger contexts to simulate model distraction across the CliniQG4QA dataset using additional models (Meditron-7B, Qwen2.5-7B-Medical, Medgemma-4B, Phi-3-mini-4k-Instruct, and GPT4o-Mini). We examine how retrieval design choices alter the accuracy–latency trade-off, examining how relevance, corpus design, and hardware constraints interact in medical retrieval-augmented generation (RAG) systems. Our comprehensive results demonstrate when retrieval is genuinely beneficial versus when it imposes unnecessary computational costs, highlighting interactions between relevance and corpus designs in QA.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030063

Authors: Ivan Malashin Dmitry Martysyuk Vadim Tynchenko Andrei Gantimurov Vladimir Nelyub Aleksei Borodulin

Mapping thematic structure in large scientific corpora enables the systematic analysis of research trends and conceptual organization. This work presents an unsupervised framework that leverages large language models (LLMs) as fixed semantic inference operators guided by structured soft prompts. The framework transforms raw abstracts into normalized semantic representations that reduce stylistic variability while retaining core conceptual content. These representations are embedded into a continuous vector space, where density-based clustering identifies latent research themes without predefining the number of topics. Cluster-level interpretation is performed using LLM-based semantic decoding to generate concise, human-readable descriptions of the discovered themes. Experiments on ICML and ACL 2025 abstracts demonstrate that the method produces coherent clusters reflecting problem formulations, methodological contributions, and empirical contexts. The findings indicate that prompt-driven semantic normalization combined with geometric analysis provides a scalable and model-agnostic approach for unsupervised thematic discovery across large scholarly corpora.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030062

Authors: Mahshid Dehghanpour Mansoor Fateh Zeynab Mohammadpoory Saideh Ferdowsi

Multiple Sclerosis (MS) and Myelitis are serious inflammatory spinal cord disorders with overlapping clinical symptoms and radiological characteristics, making accurate differentiation challenging yet clinically essential. Early and precise diagnosis is critical for guiding treatment strategies and improving patient outcomes. In this study, we propose KhayyamNet, a novel hybrid deep learning architecture designed to fuse complementary local and global representations for the accurate diagnosis of MS and Myelitis using spinal MRI. To improve robustness and generalization capability, a comprehensive preprocessing strategy including data augmentation and intensity normalization is also applied to reduce noise and address data variability. The proposed architecture combines three complementary deep learning models for feature extraction composed of Xception for high-level semantic features, Convolutional Neural Networks (CNNs) for fine-grained local patterns, and Vision Transformers (ViTs) for global contextual representations via attention mechanisms. Extracted features are then fused and refined using the Minimum Redundancy Maximum Relevance (MRMR) algorithm to eliminate redundancy and retain the most informative signals. Finally, a Random Forest (RF) classifier utilizes the optimized feature set to achieve accurate and robust differentiation between MS, Myelitis, and control spinal MRIs. Experimental results demonstrate that KhayyamNet outperforms existing methods by achieving an average classification accuracy of 98.15±0.80%. This framework demonstrates promising performance for the automated analysis of spinal MRIs and shows potential to assist in the differentiation of MS and Myelitis. While these findings highlight the potential of KhayyamNet for automated MRI interpretation, its evaluation is limited to a single-center dataset, and further validation on external multi-center data is required.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030061

Authors: Jadeyn Feng Allison Lau Melinda Hodkiewicz Caitlin Woods Michael Stewart

The creation and application of classification category labels are essential tasks for transforming complex information into structured knowledge. Categories are used for summary and reporting purposes and have historically been identified by domain experts based on their past experiences and norms. Our interest lies in the general case where expert-generated category lists require improvement, and unsupervised learning, on its own, struggles to effectively identify categories for multi-class classification of human-generated texts. We hypothesise that including an annotated knowledge graph (KG) in an embedding process will positively impact unsupervised clustering performance. Our goal is to identify clusters that can be labelled and used for classification. We look at unsupervised clustering of Maintenance Work Order (MWO) texts. MWOs capture vital observations about equipment failures in process and heavy industries. The selected KG contains a mapping of equipment types to their inherent function based on the IEC 81346-2 international standard for classification of objects in industrial systems. Performance is assessed by statistical analysis, subject matter experts, and Normalized Mutual Information score. We demonstrate that Word2Vec Bi-LSTM and Sentence-BERT NN embedding methods can leverage equipment inherent function information in the KG to improve failure mode cluster identification for the MWO. Organisations seeking to use AI to automate assignment of a failure mode code to each MWO currently need test sets classified by humans. The results of this work suggest that a semantic layer containing a knowledge graph mapping equipment types to inherent function, and inherent function to failure modes could assist in quality control for automated failure mode classification.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030060

Authors: Tea Krčmar Dina Šabanović Miljenko Švarcmajer Ivica Lukić

Anomaly detection on tabular data is widely used in fraud detection, predictive maintenance, and medical screening. While heterogeneous ensembles combining multiple detection paradigms achieve strong performance, their computational cost limits deployment in latency-sensitive or resource-constrained environments. We propose KD-AnomalyNet, a teacher–student framework that distills anomaly knowledge from a high-capacity ensemble into a lightweight neural model for efficient inference. Beyond performance replication, we study how anomaly representations transfer during distillation. To this end, we introduce a noise perturbation analysis that serves as a diagnostic probe for representation stability without introducing additional trainable components. Experiments on ten benchmark datasets show that the distilled model preserves up to 98.5% of the teacher’s AUC-ROC on the nine capacity-sufficient datasets (84.7% mean retention across all ten datasets) while achieving 26–181× inference speedups. Our analysis reveals which forms of anomaly knowledge transfer reliably—global outliers (78% transfer) and isolation-based detection (88% retention)—and which degrade under compression—local outliers (20% transfer) and neighborhood-based detection (76% retention)—providing practical guidance for deploying distilled anomaly detectors.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030059

Authors: Polydoros N. Papadopoulos Vasilis N. Burganos

This work presents a fast neural surrogate capable of reconstructing fully three-dimensional hemodynamic velocity fields in stenotic and bifurcating microvascular geometries with satisfactory accuracy, avoiding repeated, computationally demanding computational fluid dynamics (CFD) simulations. A Fourier-augmented, coordinate-neural surrogate is presented and assessed for rapid computation of three-dimensional blood-flow fields in a sample geometry. The model is trained on detailed CFD data across a parameter set of stenosis severities that feed a direct mapping from spatial coordinates to velocity components. To mitigate spectral bias and improve accuracy in regions of steep gradients, the input space is embedded with random Fourier features and compared against a conventional multilayer perceptron (MLP) backbone. Predictive ability is assessed upon strict hold-out testing, during which certain arteriolar stenosis cases are excluded from training and treated with the Fourier surrogate. Direct comparison with CFD results reveals that the Fourier MLP achieves nearly CFD fidelity with the coefficient of determination R2 ≥ 0.994 and offers more than 80% reduction in the normalized errors as provided by conventional MLP, with the precise improvement depending on the severity of stenosis. Centerline velocity and cross-sectional profiles further show that the Fourier MLP reconstructs stenosis speed-up and radial profiles more reliably compared to conventional MLP. These results indicate that Fourier feature embedding provides a simple and effective route to robust full-field hemodynamic surrogates for efficient screening of stenosis configurations without resorting to repeated, heavily demanding CFD simulations.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030058

Authors: Michael Nofer Valdrin Nimani Oliver Hinz

Hit Song Science aims to forecast a song’s success before release and benefits from integrating signals beyond audio content alone. We present DeepHits, an end-to-end multimodal network that combines (i) log-Mel spectrogram embeddings from a compact residual 2D-CNN, (ii) frozen multilingual BERT lyric embeddings, and (iii) structured numeric features including high-level Spotify audio descriptors and contextual metadata (artist popularity, release year). Evaluated on 92,517 tracks from the SpotGenTrack dataset, DeepHits achieves a macro-F1 of 52.20% (accuracy 82.63%) in the established three-class setting and a macro-F1 of 23.15% (accuracy 37.00%) in a ten-class decile benchmark. To contextualize fine-grained performance, we report capacity-controlled shallow baselines, including metadata-only and early/late fusion variants, and show that the deep multimodal model provides a clear gain over these references (e.g., metadata-only: macro-F1 20.92%; accuracy 34.22%). Ablation results indicate that removing metadata yields the largest degradation in class-balanced performance, highlighting the strong predictive value of artist popularity and release year. Overall, DeepHits provides a reproducible benchmark and modality analysis for fine-grained popularity prediction under class imbalance.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030057

Authors: Daisuke Akiba

Recent benchmarks increasingly report that large language models (LLMs) exhibit human-like causal reasoning abilities, including counterfactual inference and intervention planning. However, many such evaluations rely on domains that are heavily represented in training data and embed strong semantic cues, raising the possibility that apparent causal competence may reflect semantic pattern recombination rather than structure-sensitive causal reasoning. Drawing on human developmental theories of causal induction, this perspective argues that genuine causal understanding requires robustness to novelty and reliance on conditional structure rather than semantic familiarity. To illustrate the testability of this claim, the paper includes a pilot demonstration using synthetic causal micro-worlds. Identical numerical evidence was presented to a LLM under two conditions: semantically meaningful variable labels and non-semantic coded labels. Across paired cases, the model reliably selected the correct causal structure when labels were meaningful, but frequently misidentified or exhibited instability in causal model selection under coded labels, despite producing locally coherent explanations. These divergences emerged most clearly in diagnostically ambiguous settings requiring suppression of misleading marginal associations. The results align with the claim that semantic scaffolding can support and stabilize apparent causal competence in LLMs without implying structure-sensitive reasoning.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030056

Authors: Elena Ryumina Alexandr Axyonov Darya Koryakovskaya Timur Abdulkadirov Angelina Egorova Sergey Fedchin Alexander Zaburdaev Dmitry Ryumin

The growing demand for personalized human–computer interaction calls for methods that jointly model emotional states and personality traits. However, large-scale multimodal corpora annotated for both tasks are still lacking. This challenge stems from integrating diverse, task-specific corpora with divergent modality informativeness and domain characteristics. To address it, we propose SSL-MEPR, a semi-supervised multi-task cross-domain learning framework for Multimodal Emotion and Personality Recognition, which enables cross-task knowledge transfer without jointly labeled data. SSL-MEPR employs a three-stage strategy, progressively integrating unimodal single-task, unimodal multi-task, and multimodal multi-task models. Key innovations include Graph Attention Fusion, task-specific query-based cross-attention, predict projectors, and guide banks, which enable robust fusion and effective use of semi-labeled data via a modified GradNorm method. Evaluated on MOSEI (emotion) and FIv2 (personality), SSL-MEPR achieves a mean Weighted Accuracy (mWACC) of 70.26 and a mean Accuracy (mACC) of 92.88 in single-task cross-domain settings, outperforming state-of-the-art methods. Multi-task learning reveals domain-induced misalignment in modality informativeness but still uncovers consistent psychological patterns: sadness correlates with lower personality trait scores, while happiness aligns with higher ones. This work establishes a new paradigm for extracting cross-task psychological knowledge from disjoint multimodal corpora, demonstrating that semi-supervised multi-task cross-domain learning can bridge annotation gaps while preserving theoretically grounded emotion–personality relationships.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030055

Authors: Ismail Lamaakal Chaymae Yahyati Yassine Maleh Khalid El Makkaoui Ibrahim Ouahbi

Human Activity Recognition (HAR) on wearable and IoT devices must jointly satisfy four requirements: high accuracy, the ability to recognize previously unseen activities, strict memory and latency constraints, and interpretable decisions. In this work, we address all four by introducing an explainable Kolmogorov–Arnold Network for Human Activity Recognition (TinyKAN-HAR) with a zero-shot learning (ZSL) module, designed specifically for TinyML edge devices. The proposed KAN replaces fixed activation functions by learnable one-dimensional spline operators applied after linear mixing, yielding compact yet expressive feature extractors whose internal nonlinearities can be directly visualized. On top of the KAN latent space, we learn a semantic projection and cosine-based compatibility function that align sensor features with class-level semantic embeddings, enabling both pure and generalized zero-shot recognition of unseen activities. We evaluate our method on three benchmark datasets (UCI HAR, WISDM, PAMAP2) under subject-disjoint and zero-shot splits. TinyKAN-HAR consistently achieves over 97% macro-F1 on seen classes and over 96% accuracy on unseen activities, with harmonic mean above 96% in the generalized ZSL setting, outperforming CNN, LSTM and Transformer-based ZSL baselines. For explainability, we combine gradient-based attributions, SHAP-style global relevance scores and inspection of the learned spline functions to provide sensor-level, temporal and neuron-level insights into each prediction. After 8-bit quantization and TinyML-oriented optimizations, the deployed model occupies only 145 kB of flash and 26 kB of RAM, and achieves an average inference latency of 4.1 ms (about 0.32 mJ per window) on a Cortex-M4F-class microcontroller, while preserving accuracy within 0.2% of the full-precision model. These results demonstrate that explainable, zero-shot HAR with near state-of-the-art accuracy is feasible on severely resource-constrained TinyML edge devices.

Machine Learning and Knowledge Extraction doi: 10.3390/make8030054

Authors: Abdulrazaq Nafiu Abubakar Ali Nasir Abdul-Wahid A. Saif

Modern multi-Unmanned Aerial Vehicle (UAV) attacks pose significant challenges to existing counter-UAV frameworks due to their agility, irregular spatial formations, and increasing reliance on intelligent evasive behaviors. This paper proposes a unified interception architecture that integrates Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for multi-target grouping, a deceptive waypoint sequencing (DWS) mechanism for adversarial evasion, and a robust sliding-mode backstepping controller augmented with extended state observers (ESOs) for precise tracking under disturbances. DBSCAN enables real-time clustering of attacking UAVs without prior knowledge of the number of formations, producing dynamic centroids that serve as tactical interception references. To counter risky attackers capable of predicting defender trajectories, a novel DWS strategy introduces centroid-relative waypoints that preserve mission objectives while reducing trajectory predictability. Lyapunov-based analysis is developed for stability, guaranteeing uniform ultimate boundedness of the tracking errors. The proposed approach achieves successful interception in both scenarios, with an interception time of 7 s and final interception error of 0.023 m in the single-UAV case, and an interception time of 8 s with final interception error of 0.050 m in the multiple-UAV case, whereas the PID baseline fails to achieve interception under the same conditions. Extensive simulations involving single and multi-cluster engagements demonstrate that the proposed strategy achieves fast, accurate, and deception-resilient interception, outperforming the conventional PID approach in the presence of disturbances, nonlinearities, and dynamic swarm configurations. The obtained results show the effectiveness of integrating adaptive clustering, deceptive planning, and robust nonlinear control for modern UAV–UAV defensive operations.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020053

Authors: Hesham Kamal Maggie Mashaly

As network environments become increasingly interconnected, ensuring robust cyber-security has become critical, particularly with the growing sophistication of modern cyber threats. Intrusion detection systems (IDSs) play a vital role in identifying and mitigating unauthorized or malicious activities; however, conventional machine learning-based IDSs often rely on handcrafted features and are limited in their ability to detect diverse attack types across disparate network domains. To address these limitations, this paper introduces a novel unified intrusion detection framework that implements “Structural Dualism” to integrate three heterogeneous benchmark datasets (CSE-CIC-IDS2018, NF-BoT-IoT-v2, and IoT-23) into a harmonized, protocol-agnostic representation. The framework employs a shared autoencoder architecture with dataset-specific projection layers to learn a unified latent manifold. This 15-dimensional space captures the underlying semantics of attack patterns (e.g., volumetric vs. signaling) across multiple domains, while dataset-specific decoders preserve reconstruction fidelity through alternating multi-domain training. To identify complex micro-signatures within this manifold, the framework utilizes a synergistic hybrid convolutional neural network–deep neural network (CNN–DNN) classifier, where the CNN extracts spatial latent patterns and the DNN performs global classification across twenty-five distinct classes. Class imbalance is addressed through resampling strategies such as adaptive synthetic sampling (ADASYN) and edited nearest neighbors (ENN). Experimental results demonstrate remarkable performance, achieving 99.76% accuracy for binary classification and 99.54% accuracy for multi-class classification on the merged dataset, with strong generalization confirmed on individual datasets. These findings indicate that the shared autoencoder-based CNN–DNN framework, through its unique feature alignment and spatial extraction capabilities, significantly strengthens intrusion detection across diverse and heterogeneous environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020052

Authors: Lakhdar Remaki

Support Vector Machine (SVM) is a popular kernel-based method for data classification that has demonstrated high efficiency across a wide range of practical applications. However, SVM suffers from several limitations, including the potential failure of the optimization process, especially in high-dimensional spaces; the inherently high computational cost; the lack of a systematic approach to multi-class classification; difficulties in handling imbalanced classes; and the prohibitive cost of real-time or dynamic classification. This paper proposes an alternative method, referred to as Kernel-based Optimal Subspaces (KOS), which belongs to the family of kernel subspace methods. Mathematically similar to Kernel PCA (KPCA), KOS achieves performance comparable to SVM while addressing the aforementioned weaknesses. The method is based on computing the minimum distance to optimal feature subspaces of the mapped data. Because no optimization process is required, KOS is robust, fast, and easy to implement. The optimal subspaces are constructed independently, enabling high parallelizability and making the approach well-suited for dynamic classification and real-time applications. Furthermore, the issue of imbalanced classes is naturally handled by subdividing large classes into smaller sub-classes, thereby creating appropriately sized sub-subspaces within the feature space.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020051

Authors: Nicola Furnitto Giuseppe Todde Maria Spagnuolo Giuseppe Sottosanti Maria Caria Giampaolo Schillaci Sabina I. G. Failla

In recent years, agriculture has begun to transform thanks to the arrival of robots and autonomous vehicles capable of performing complex operations such as weeding and spraying in an intelligent and targeted manner. In fact, new-generation agricultural robots use artificial intelligence (AI), cameras, and sensors to recognise weeds, analyse crop conditions, and apply plant protection products only where necessary, thus reducing waste and environmental impact. Some systems combine drones and ground vehicles to achieve even more accurate results. This systematic review synthesises recent advances in agricultural robotics for weed and pest management through a PRISMA-based approach. Literature was collected from major scientific databases (Scopus, Web of Science, IEEE Xplore, Google Scholar) and complementary sources, leading to the inclusion of 83 eligible studies. The selected evidence was structured into four application domains: (i) weed detection and mapping, (ii) robotic and non-chemical weed control (mechanical and laser-based approaches), (iii) selective/variable-rate spraying for pest and disease management, and (iv) integrated weeding–spraying solutions, including cooperative Unmanned Aerial Vehicle–Unmanned Ground Vehicle (UAV–UGV) systems. Overall, the reviewed studies confirm rapid progress in real-time perception (deep learning-based detection), navigation/localization (e.g., GNSS/RTK, LiDAR, sensor fusion) and targeted actuation (spot spraying and precision interventions), while also revealing persistent limitations: heterogeneous evaluation protocols, limited system-level comparisons in terms of work rate, scalability, costs and robustness under variable field conditions, and an often unclear distinction between prototype platforms and solutions close to commercialization. However, the large-scale spread of these technologies is still hampered by high costs, technical complexity, and cultural resistance. The review highlights how the integration of automation, sustainability, and accessibility is key to the agriculture of the future.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020050

Authors: Nor Saiful Azam Bin Nor Azmi Michal Ptaszynski Fumito Masui Abu Nowhash Chowdhury

The rise of toxic content, particularly in the form of hateful memes, poses a significant challenge to social media platforms. This paper presents an empirical comparative study of unimodal and multimodal architectures for toxic content detection. Rather than proposing a novel architecture, the study evaluates the efficacy of a modular Late Fusion framework (RoBERViT) against specialized unimodal baselines (RoBERTa and ViT) and a generalist Large Multimodal (LLaVA). Both unimodal and multimodal configurations across two distinct benchmarks—the imbalanced Innopolis Hateful Memes dataset and the confounder-driven Facebook Hateful Meme dataset—were explored. Beyond quantitative metrics, this study conducts a qualitative analysis using Explainable AI (LIME) and a Large Multimodal Model (LLaVA) to investigate model reasoning. Results demonstrate that the multimodal fusion model consistently outperformed its unimodal counterparts on the Innopolis Hateful Meme dataset, achieving a toxic class F1-score of 0.6439 compared to the text-only score of 0.5794. However, on the Facebook Hateful Meme dataset, text-only models remain competitive, highlighting the “benign confounder” challenge. The qualitative analysis reveals that text remains the dominant modality, with models often relying on surface-level keywords. Notably, the Vision Transformer frequently uses text overlays as a visual proxy for hate, while the LLaVA model struggles with hallucinated toxicity in benign confounder contexts. These findings underscore the persistent challenge of achieving true multimodal understanding in hate speech detection.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020049

Authors: Sebastian Rojas-Ordoñez Mikel Segura Irune Yarza Veronica Mendoza Ekaitz Zulueta

Large Language Models are increasingly used for high-level robotic reasoning, yet their latency and stochasticity complicate their direct use in low-level control. Moreover, extracting actionable navigation cues from multimodal context incurs inference costs that are challenging for embedded platforms. We present a plug-and-play framework that augments a finite-state machine with asynchronous velocity suggestions generated by a Large Language Model, using an off-the-shelf DistilGPT-2 model running on-device on a Jetson AGX Orin. The system extracts task-relevant cues from the current context and integrates them only if they satisfy deadline, schema, and kinematic validation, thereby preserving a deterministic 50 Hz control loop with a <5 ms fallback path. We compare multiple Large Language Models for embedded robot control and quantify trade-offs among model size, inference time, and output validity. To assess whether the Large Language Models add value beyond signal processing, we include an ablation against a standard smoothing baseline; the results indicate that the Large Language Models contribute anticipatory, context-dependent adjustments that are not captured by filtering alone. Experiments in Gazebo and on a real TurtleBot3 reduce the final position error from 0.246 m to 0.159 m and improve trajectory efficiency from 0.821 to 0.901 without increasing control-loop latency. Approximately 80% of the Large Language Models’ outputs pass validation and are applied. Overall, the framework reduces developer effort by enabling behavioral changes at the prompt level while maintaining interpretable, robust edge-based navigation.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020047

Authors: Sara Nassar Rachid Hedjam Samir Brahim Belhaouari

A popular method for projecting high-dimensional data onto a lower-dimensional space while preserving the integrity of its structure is t-distributed Stochastic Neighbor Embedding (t-SNE). This technique minimizes the Kullback–Leibler (KL) divergence to align the similarities between points in the original and reduced spaces. While t-SNE is highly effective, it prioritizes local neighborhood preservation, which results in limited separation between distant clusters and inadequate representation of global relationships. To improve these limitations, this work introduces two complementary approaches: (1) The Max-Flipped KL Divergence (KLmax) modifies the original divergence by incorporating a contrastive term, KL′, which enhances the ranking of point similarities through maximum similarity constraints. (2) The KL-Wasserstein Loss (LKL−W) combines the KL divergence with the classic Wasserstein distance, allowing the embedding to benefit from the smooth and geometry-aware transport properties of Wasserstein metrics. Experimental results show that these methods lead to improved separation and better structural clarity in the low-dimensional space compared to standard t-SNE.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020048

Authors: Mustafa Abdulkadhim Sandor R. Repas

The transition of Large Language Models (LLMs) from centralized clouds to edge environments is critical for addressing privacy concerns, latency bottlenecks, and operational costs. However, existing edge benchmarking frameworks remain tailored to discriminative Deep Learning tasks (e.g., object detection), failing to capture the multidimensional challenges of generative AI, specifically the trade-offs between token generation speed, semantic accuracy, and hardware sustainability. To address this gap, we introduce LEAF (LLM Edge Assessment Framework), a novel evaluation methodology that integrates Circular Economy principles directly into performance metrics. LEAF assesses edge deployments across five synergistic pillars: Circular Economy Score, Energy Efficiency (Joules/Token), Performance Speed (Tokens/Second), semantic accuracy (BERTScore), and End-to-End Latency. We validate LEAF through an extensive experimental analysis of five distinct hardware classes, ranging from embedded IoT devices (Raspberry Pi 4 and 5, NVIDIA Jetson Nano) to professional edge servers (NVIDIA T400) and repurposed legacy workstations (NVIDIA GTX 1050 Ti). Utilizing 4-bit quantized models via the Ollama runtime, our results reveal a counterintuitive insight: repurposed consumer hardware significantly outperforms modern purpose-built edge SoCs. The legacy GTX 1050 Ti achieved a 20× speedup over the Raspberry Pi 4 and maintained superior energy-per-task efficiency compared to low-power ARM architectures by minimizing active runtime. These findings challenge the prevailing narrative that newer silicon is essential for Edge AI, demonstrating that sustainable, high-performance inference can be achieved by extending the lifecycle of existing hardware. LEAF thus provides a blueprint for a “Green Edge” ecosystem that balances computational capability with environmental responsibility.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020046

Authors: Haiyang Gu Yong Wang Huaiyang Liu Tong Tian Changxing Geng Yun Shi

Perception in trellised orchards is often challenged by dense canopy occlusion and overhead plastic coverings, which cause pronounced variations in sky visibility at row terminals. Accurately recognizing row terminals, including both row head and row tail positions, is therefore essential for understanding orchard row structures. This study presents SkySeg-Net, a sky segmentation-based framework for row-terminal recognition in trellised orchards. SkySeg-Net is built on an enhanced multi-scale U-Net architecture and employs ResNeSt residual split-attention blocks as the backbone. To improve feature discrimination under complex illumination and occlusion conditions, the Convolutional Block Attention Module (CBAM) is integrated into the downsampling path, while a Pyramid Pooling Module (PPM) is introduced during upsampling to strengthen multi-scale contextual representation. Sky regions are segmented from both front-view and rear-view camera images, and a hierarchical threshold-based pixel-sum analysis is applied to infer row-terminal locations based on sky-region distribution patterns. To support a comprehensive evaluation, a dedicated trellised vineyard dataset was constructed, featuring front-view and rear-view images and covering three representative grapevine growth stages (BBCH 69–71, 73–77, and 79–89). Experimental results show that SkySeg-Net achieves an mIoU of 91.21% and an mPA of 94.82% for sky segmentation, with a row-terminal recognition accuracy exceeding 98.17% across all growth stages. These results demonstrate that SkySeg-Net provides a robust and reliable visual perception approach for row-terminal recognition in trellised orchard environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020045

Authors: Wei Dong Han Zhou Terry Ji Jun Chen

Adverse weather removal aims to restore images degraded by haze, rain, or snow. However, existing unified models often rely on implicit degradation cues, making them vulnerable to inaccurate weather perception and insufficient semantic guidance, which leads to over-smoothing or residual artifacts in real scenes. In this work, we propose AWR-VIP, a prior-guided adverse weather removal framework that explicitly extracts semantic and perceptual priors using a frozen vision–language model (VLM). Given a degraded input, we first employ a degradation-aware prompt extractor to produce a compact set of semantic tags describing key objects and regions, and simultaneously perform weather-type perception by prompting the VLM with explicit weather definitions. Conditioned on the predicted weather type and selected tags, the VLM further generates two levels of restoration guidance: a global instruction that summarizes image-level enhancement goals (e.g., visibility/contrast) and local instructions that specify tag-aware refinement cues (e.g., recover textures for specific regions). These textual outputs are encoded by a text encoder into a pair of priors (Pglobal and Plocal), which are injected into a UNet-based restorer through global-prior-modulated normalization and instruction-guided attention, enabling weather-adaptive and content-aware restoration. Extensive experiments on a combined benchmark show that AWR-VIP consistently outperforms state-of-the-art methods. Moreover, the VLM-derived priors are plug-and-play and can be integrated into other restoration backbones to further improve performance.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020044

Authors: Atif Latif Masoumeh Hashemi Matt Yost Somayeh Esmaeili Xiaojun Qi

Semantic segmentation and deep learning methods have rarely been applied to fractional vegetation cover (FVC) segmentation tasks due to the lack of publicly available datasets for training deep learning models. FVC is a key indicator for assessing vegetation distribution, crop density, and crop responses to water availability and fertilizer application, yet conventional field-based measurement methods are time consuming, costly, labor intensive, and may lack the accuracy required for critical applications such as drought stress evaluation and water productivity. In this paper, we introduced causality-based deep learning techniques for FVC segmentation on a publicly available RGB dataset that consists of four ground cover crops: Phyla nodiflora L., Cynodon dactylon, Frankenia thymifolia Desf., and Oxalis stricta L. By separating causal from spurious correlations in pretrained features, using the stepwise intervention and reweighting (SIR) method at different encoder stages reduced confounding bias and enabled the models to learn more generalizable and task-relevant features. Extensive experiments on the FVC dataset, conducted with and without causality learning, showed that the proposed FCN + ResNet-50 model with causality learning and data augmentation achieved an accuracy of 94.80%, a precision of 94.97%, a recall of 94.35%, and an F1-score of 94.62%, which outperformed non-causal baselines and state-of-the-art transformer-based models including SegFormer and Mask2Former.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020043

Authors: Aristeidis Karras Leonidas Theodorakopoulos Christos Karras Alexandra Theodoropoulou

Modern Autonomous Vehicles generate large volumes of heterogeneous in-vehicle data, making cybersecurity a critical challenge as adversarial attacks become increasingly adaptive, stealthy, and multi-protocol. Traditional intrusion detection systems often fail under these conditions because of their limited contextual understanding, poor robustness to distribution shifts, and insufficient regulatory transparency. This study introduces LLM-Guardian, a hierarchical intrusion detection framework with decision-making mechanisms that integrates Large Language Models (LLMs) with classical statistical detection theory, optimal transport drift analysis, graph neural networks, and formal uncertainty quantification. LLM-Guardian uses semantic anomaly scoring, conformal prediction for distribution-free confidence calibration, adaptive cumulative sum (CUSUM) sequential testing for low-latency detection, and topology-aware GNN reasoning designed to identify coordinated attacks across CAN, Ethernet, and V2X interfaces. In this work, the framework is empirically evaluated on four heterogeneous CAN-bus datasets, while the Ethernet and V2X components are instantiated at the architectural level and left as directions for future multi-protocol experimentation.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020042

Authors: Philipp Brunner Stefanie Vogl

This paper presents a scalable machine learning pipeline for extracting actionable, product-related insights from user-generated social media comments. Leveraging sentence embeddings from SBERT and unsupervised clustering (k-Means and agglomerative), the approach structures informal and noisy comments from Instagram and YouTube into topic groups intended to support thematic analysis. A case study on feedback regarding BMW vehicles, comprising more than 26,000 comments, illustrates how the pipeline can reveal recurring user concerns, such as design critiques, usability issues, and technology-related expectations, even in short and unstructured social media comments. The proposed pipeline operates without labeled data or manual annotation, enabling scalable application and transferability across product categories and industries. By transforming large-scale, unstructured consumer feedback into interpretable themes, the pipeline provides product teams with an efficient and structured basis for data-driven product development and improvement.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020041

Authors: Somaiya Al Shuraiqi Adhari AlZaabi Abdulrahman Aal Abdulsalam

The use of large language models (LLMs) to automate the generation of medical case-based multiple-choice questions (MCQs) is increasing, but their accuracy, reliability, and educational validity are still not well understood. This study in a comparative framework examined nine LLMs with four different prompting methods to evaluate LLM-produced MCQs for clinical coherence and readiness for assessment. A uniform evaluation pipeline was constructed to examine automatic text-similarity measures using automated metrics (BLEU, ROUGE, and METEOR), structural and parsability measures, and operational effectiveness (latency, cost, quality-efficiency ratios). Human validation was performed on the best-performing model and prompt combination (OpenBioLLM-70B with Chain-of-Thought) focusing on the model prompt that demonstrated the best linguistic fidelity and clinically aligned reasoning. Two clinical experts independently reviewed 88 items using a five-domain rubric covering appropriateness, clarity, relevance, distractor quality, and cognitive level. Results indicated significant variation across models and prompting strategies, with Chain-of-Thought yielding the best overall performance in comparison to other strategies. The OpenBioLLM-70B model demonstrated the best overall balance of quality, parsability, and efficiency, achieving a prompt template quality score of 90.4, a consistency score of 88.8, and a response time of 3.28 s, with a quality-per-dollar value of 134.11. The expert rating confirmed clinical alignment, but there was consensus that distractor quality needed further improvements. These results provide evidence that LLMs under optimal prompting conditions can reliably support MCQ generation and provide large-scale, cost-effective support for medical assessment production.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020040

Authors: Omneya Attallah

The early and accurate identification of breast cancer is a significant healthcare issue, largely because the traditional machine learning approaches rely on handcrafted features that are unable to fully capture the spatial and textural complexity found in mammograms. Even with the advancements made possible through deep learning and improvements in diagnostic performance, most computational-aided diagnosis (CAD) systems based on Convolutional Neural Networks (CNNs) still only rely on single-domain features, normally spatial features, while neglecting some important spectral and spatial–spectral features, leading to limitations in generalisability, redundancy, and loss of performative interpretability. Inspired by these limitations, this research proposes MERGE, a novel CAD framework that combines spatial, spectral, and spatial–spectral information—all part of a single multistage architecture taking advantage of three fine-tuned CNN models (ResNet-50, Xception, and Inception). This system utilises Discrete Stationary Wavelet Transform (DSWT) to enhance spectral–spatial features; Discrete Cosine Transform (DCT) to fuse the features optimally, resulting in enhanced spatial and spatial–spectral representations; and, finally, Non-Negative Matrix Factorisation (NNMF) for reduced-dimensional features. Finally, the Linear Discriminant Analysis (LDA), support vector machine (SVM), and k-nearest neighbours (KNN) classifiers provide a robust diagnosis. Using the INBreast and MIAS datasets in evaluations of the experimental research design, evaluation metrics of accuracy, sensitivity, specificity, and AUC were around 99%, with performance surpassing state-of-the-art paradigms. The findings of the suggested MERGE indicate significant promise as a dependable and effective diagnostic tool, enhancing the consistency and interpretability of breast cancer screening results.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020039

Authors: Renyuan Shen Yong Wang Huaiyang Liu Haiyang Gu Changxing Geng Yun Shi

Dense foliage, severe illumination variations, and interference from multiple individuals with similar appearances in complex orchard environments pose significant challenges for vision-based following robots in maintaining persistent target perception and identity consistency, thereby compromising the stability and safety of fruit transportation operations. To address these challenges, we propose a novel framework, DeepDIMP-ReID, which integrates the Deep Implicit Model Prediction (DIMP) tracker with a person re-identification (ReID) module based on EfficientNet. This visual perception and autonomous following framework is designed for differential-drive orchard transportation robots, aiming to achieve robust target perception and reliable identity maintenance in unstructured orchard settings. The proposed framework adopts a hierarchical perception–verification–control architecture. Visual tracking and three-dimensional localization are jointly achieved using synchronized color and depth data acquired from a RealSense camera, where target regions are obtained via the discriminative model prediction (DIMP) method and refined through an elliptical-mask-based depth matching strategy. Front obstacle detection is performed using DBSCAN-based point cloud clustering techniques. To suppress erroneous following caused by occlusion, target switching, or target reappearance after occlusion, an enhanced HOReID person re-identification module with an EfficientNet backbone is integrated for identity verification at critical decision points. Based on the verified perception results, a state-driven motion control strategy is employed to ensure safe and continuous autonomous following. Extensive long-term experiments conducted in real orchard environments demonstrate that the proposed system achieves a correct tracking rate exceeding 94% under varying human walking speeds, with an average localization error of 0.071 m. In scenarios triggering re-identification, a target discrimination success rate of 93.3% is obtained. These results confirm the effectiveness and robustness of the proposed framework for autonomous fruit transportation in complex orchard environments.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020038

Authors: Sebastiano Gangemi Alessandro Allegra Mario Di Gioacchino Luca Gammeri Irene Cacciola Giorgio Walter Canonica

Guidelines provide specific recommendations based on the best available medical knowledge, summarizing and balancing the advantages and disadvantages of various diagnostic and treatment options. Currently, consensus methods are the best and most common practices in creating clinical guidelines, even though these approaches have several limitations. However, the rapid pace of biomedical innovation and the growing availability of real-world data (RWD) from clinical registries (containing data like clinical outcomes, treatment variables, imaging, and laboratory results) call for a complementary paradigm in which recommendations are continuously stress-tested against high-quality, interoperable data and auditable artificial intelligence (AI) pipelines. AI, based on information retrieved from patient registries, can optimize the process of creating guidelines. In fact, AI can analyze large volumes of data, ensuring essential tasks such as correct feature identification, prediction, classification, and pattern recognition of all information. In this work, we propose a four-phase lifecycle, comprising data curation, causal analysis and estimation, objective validation, and real-time updates, complemented by governance and machine learning operations (MLOps). A comparative analysis with consensus-only methods, a pilot protocol, and a compliance checklist are provided. We believe that the use of AI will be a valuable support in drafting clinical guidelines to complement expert consensus and ensure continuous updates to standards, providing a higher level of evidence. The integration of AI with high-quality patient registries has the potential to substantially modernize guideline development, enabling continuously updated, data-driven recommendations.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020037

Authors: Lars Wilhelmi Christian Bruns Matthias Schumann

This study investigates how Vision Language Models (VLMs) can be used and methodically configured to extract Environmental, Social, and Governance (ESG) metrics from corporate sustainability reports, addressing the limitations of existing text-only and manual ESG data-extraction approaches. Using the Design Science Research Methodology, we developed an extraction artifact comprising a curated page-level dataset containing greenhouse gas (GHG) emission-reduction targets, an automated evaluation pipeline, model and text-preprocessing comparisons, and iterative prompt and few-shot refinement. Pages from oil and gas sustainability reports were processed directly by VLMs to preserve visual–textual structure, enabling a controlled comparison of text, image, and combined input modalities, with extraction quality assessed at page and attribute level using F1-scores. Among tested models, Mistral Small 3.2 demonstrated the most stable performance and was used to evaluate image, text, and combined modalities. Combined text + image modality performed best (F1 = 0.82), particularly on complex page layouts. The findings demonstrate how to effectively integrate visual and textual cues for ESG metric extraction with VLMs, though challenges remain for visually dense layouts and avoiding inference-based hallucinations.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020036

Authors: Frida Milella Federico Cabitza

The increasing integration of artificial intelligence (AI) in decision-making processes has amplified discussions surrounding algorithmic authority—the perceived epistemic legitimacy of AI systems over human judgment. This study investigates how individuals attribute epistemic authority to AI, focusing on psychological, contextual, and sociotechnical factors. Existing research highlights the importance of trust in automation, perceived performance, and moral frameworks in shaping such attributions. Unlike prior conceptual or philosophical accounts of algorithmic authority, our study adopts a relational and empirically grounded perspective by operationalizing algority through psychometric measures and contextual assessments. To address knowledge gaps in the micro-level dynamics of this phenomenon, we conducted an empirical study using psychometric tools and scenario-based assessments. Here, we report key findings from a survey of 610 participants, revealing significant correlations between trust in automation (TiA), perceptions of automated performance (PAS), and the propensity to defer to AI, particularly in high-stakes scenarios like criminal justice and job-matching. Trust in automation emerged as a primary factor, while moral attitudes moderated deference in ethically sensitive contexts. Our findings highlight the practical relevance of transparency and explainability for supporting critical engagement with AI outputs and for informing the design of contextually appropriate decision support. This study contributes to understanding algorithmic authority as a multidimensional construct, offering empirically grounded insights for designing AI systems that are trustworthy and context-sensitive.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020035

Authors: Tuba Nur Subasi Abdulhamit Subasi

Migraine and epilepsy are common neurological disorders that share overlapping symptoms, such as visual disturbances and altered consciousness, making accurate diagnosis challenging. Although their underlying mechanisms differ, both conditions involve recurrent irregular brain activity, and traditional EEG-based diagnosis relies heavily on clinical interpretation, which may be subjective and insufficient for clear differentiation. To address this challenge, this study introduces an automated EEG classification framework combining Dual Tree Complex Wavelet Transform (DTCWT) for feature extraction with a Random Subspace Ensemble Classifier for multi-class discrimination. EEG data recorded under photic and nonphotic stimulation were analyzed to capture both temporal and frequency characteristics. DTCWT proved effective in modeling the non-stationary nature of EEG signals and extracting condition-specific features, while the ensemble classifier improved generalization by training multiple models on diverse feature subsets. The proposed system achieved an average accuracy of 99.50%, along with strong F-measure, AUC, and Kappa scores. Notably, although previous studies suggest heightened EEG activity in migraine patients during flash stimulation, findings here indicate that flash stimulation alone does not reliably distinguish migraine from epilepsy. Overall, this research highlights the promise of advanced signal processing and machine learning techniques in enhancing diagnostic precision for complex neurological disorders.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020034

Authors: Mohamed A. Abdelhamed Hana M. Nassef Sara Abdelnasser Sahar Selim Lobna A. Said

Accurate lung tumor segmentation using computed tomography (CT) scans is needed for efficient tumor treatment. However, the development of deep learning models is often constrained by strict patient privacy regulations that limit direct data sharing. This work presents a system that enables multi-institutional collaboration while training high-quality lung tumor segmentation models without requiring access to sensitive patient data. The proposed framework features the AuraViT suite, which includes the standard AuraViT—a hybrid model with 136 million parameters that combines a Vision Transformer (ViT) encoder, Atrous Spatial Pyramid Pooling (ASPP), and attention-gated residual connections—and the Lightweight AuraViT (LAURA) family (Small, Tiny, and Mobile). These variants are designed for resource-constrained environments and potential edge deployment scenarios. Training is conducted on publicly available datasets (MSD Lung and NSCLC) in a simulated five-client federated learning setup that emulates collaboration among institutions while ensuring patient privacy. The framework uses a federated learning setup with FedProx, adaptive weighted aggregation, and a dynamic virtual client strategy to handle data and system differences. The framework is further evaluated through ablation studies on model architecture and feature importance. The results show that the standard AuraViT-FL achieves a global mean Dice score of 80.81%, while maintaining performance close to centralized training. Additionally, the LAURA variations show a better trade-off between accuracy and efficiency. Notably, the Mobile variant with ∼5 M parameters reduces model complexity by over 96% while maintaining competitive performance (82.96% Dice on MSD Lung).

Machine Learning and Knowledge Extraction doi: 10.3390/make8020033

Authors: Hemn Barzan Abdalla Awder Ahmed

In an e-learning platform, information retrieval plays an enormous role through efficient processing. Recently, the education sector has increased its trend in online learning systems by generating a large amount of educational content based on student’s criteria. For this sophisticated data analysis scheme, several methods have been employed in recent studies; however, they have suffered from various limitations, including reliability issues, security problems, unauthorized disclosure of data, cost consumption, and interpretability challenges. To tackle these issues, a proposed framework, named the war strategy optimization-based bidirectional long short-term memory (WSO-BiLSTM) model, is designed in this research to reduce sensitivity to local optima and improve convergence stability, thereby achieving robust retrieval performance. With this perspective, the BiLSTM model captures the semantic information of documents in a dual direction for effective retrieval outcomes. Moreover, the model’s key features are extracted effectively by various feature extraction methods. The dynamic movement towards the optimal solution of the WSO algorithm enables the proposed model to retrieve the information more accurately in the information retrieval system. Experiments on an e-learning dataset show that, with a 90% training split, the proposed method achieves 97.90% accuracy, 98.45% precision, 97.90% F1-score, and 97.35% recall.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020032

Authors: Marek Socha Agata Durawa Małgorzata Jelito Katarzyna Dziadziuszko Witold Rzyman Edyta Szurowska Joanna Polanska

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide, and emphysema is present in the majority of affected patients and can be identified on computed tomography (CT). This study investigated whether radiomic features derived from automatically and adaptively segmented low-attenuation lung regions can capture distinct imaging characteristics of COPD beyond conventional emphysema measures. Radiomic features were extracted from 6078 chest CT scans of 2243 participants from the COPDGene cohort. Emphysematous regions were segmented using the MimSeg method based on Gaussian mixture modelling with patient-adjusted thresholding, and radiomic features were computed for individual lesion clusters and aggregated per patient using summary statistics, yielding 780 features per subject. Uniform Manifold Approximation and Projection (UMAP) was used to generate a low-dimensional embedding, and feature contributions were evaluated using SHAP analysis and statistical testing. The resulting embedding demonstrated structured patterns broadly aligned with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages, with greater overlap among GOLD 0–2 and more consolidated groupings for GOLD 3 and 4, reflecting differences in disease severity. The most influential features were predominantly derived from Grey Level Run Length Matrix measures, capturing textural heterogeneity and spatial organisation of emphysematous changes that are not directly described by standard density-based metrics. These findings suggest that radiomic analysis of adaptively segmented CT data may provide complementary and structurally distinct information relative to conventional emphysema measures, supporting a more nuanced characterisation of emphysema patterns in COPD.

Machine Learning and Knowledge Extraction doi: 10.3390/make8020031

Authors: Shirin Shila Md. Safayat Hossain Md Fuyad Al Masud Mohammad Badrul Alam Miah Afrig Aminuddin Zia Muhammad

The automated and accurate results of classifying histopathology images are necessary in the early detection of cancer, especially the common cancers such as Colorectal Cancer (CRC) and Lung Cancer (LC). Nonetheless, classical deep learning frameworks often face challenges because the intra-class variations are large, the relations across classes are alike, and the quality of images is not stable. In order to eliminate these constraints, a multi-layer diagnostic framework is offered in detail. This process starts with a strong preprocessing pipeline, which involves gamma correction, bilateral filtering, and adaptive CLAHE, resulting in statistically significant changes in image quality quantitative measures. The hybrid attention architecture is presented and includes an Xception backbone, a Convolutional Block Attention Module (CBAM), a Transformer block, and an MLP classifier to successfully combine local features with global context. The proposed model achieved an outstanding performance with a classification of 99.98%, 99.58%, and 99.33% percent on LC25000, CRC-VAL-HE-7K, and NCT-CRC-HE-100K when tested on three publicly available datasets. In order to enhance transparency, very detailed explainability analyses are conducted with the help of layer-wise feature visualization and Grad-CAM. Finally, the real-world example of this framework is presented by its implementation in a web-based platform, which can be a useful and easy-to-use tool in helping to diagnose a pathology.