Machine Learning and Knowledge Extraction

With the rapid development of applications such as edge computing, the Internet of Things (IoT), and embodied intelligence, massive multimodal data are continuously generated on end devices in a streaming manner. To maintain model adaptability and robustness in dynamic environments, incremental learning has gradually become the core training paradigm on edge devices. However, edge devices are constrained by limited computational, storage, and communication resources, making it infeasible to retain and process all data samples over time. This necessitates efficient data selection strategies to reduce redundancy and improve training efficiency. Existing sample selection methods primarily focus on overall sample difficulty or gradient contribution, but they overlook the heterogeneity of multimodal data in terms of information content and discriminative power. This often leads to modality imbalance, causing the model to over-rely on a single modality and suffer performance degradation. To address this issue, this paper proposes a multimodal sample selection strategy based on the Modality Balance Score (MBS). The method computes confidence scores at the modality level for each sample and further quantifies the contribution differences across modalities. In the selection process, samples with balanced modality contributions are prioritized, thereby improving training efficiency while alleviating modality bias. Experiments conducted on two benchmark datasets, CREMA-D and AVE, demonstrate that compared with existing approaches, the MBS strategy achieves the most stable performance under medium-to-high selection ratios (0.25–0.4), yielding superior results in both accuracy and robustness. These findings validate the effectiveness of the proposed strategy in resource-constrained scenarios, providing both theoretical insights and practical guidance for multimodal sample selection in learning tasks. Full article

Background: Pituitary neuroendocrine tumours (PitNETs) are clinically and biologically heterogeneous neoplasms that remain challenging to diagnose, prognosticate, and treat. Although recent WHO classifications using transcription-factor-based markers have refined pathological categorisation, histopathology alone still fails to predict tumour behaviour or support individualised therapy. Objective: This systematic review aimed to evaluate how machine learning (ML) and knowledge extraction approaches can complement pathology by integrating multi-dimensional omics datasets to generate predictive and clinically meaningful insights in PitNETs. Methods: The review followed the PRISMA 2020 statement for systematic reviews. Searches were conducted in PubMed, Google Scholar, arXiv, and SciSpace up to June 2025 to identify omics studies applying ML or computational data integration in PitNETs. Eligible studies included original research using genomic, transcriptomic, epigenomic, proteomic, or liquid biopsy data. Data extraction covered study design, ML methodology, data accessibility, and clinical annotation. Study quality and validation strategies were also assessed. Results: A total of 726 records were identified. After the reviewing process, 98 studies met inclusion criteria. PitNET research employed unsupervised clustering or regularised regression methods reflecting their suitability for high-dimensional omics datasets and the limited sample sizes. In contrast, deep learning approaches were rarely implemented, primarily due to the scarcity of large, clinically annotated cohorts required to train such models effectively. To support future research and model development, we compiled a comprehensive catalogue of all publicly available PitNET omics resources, facilitating reuse, methodological benchmarking, and integrative analyses. Conclusions: Although omics research in PitNETs is increasing, the lack of standardised, clinically annotated datasets remains a major obstacle to the development and deployment of robust predictive models. Coordinated efforts in data sharing and clinical harmonisation are required to unlock its full potential. Full article

Background: Mosquito-borne viral diseases are a growing global health threat, and artificial intelligence (AI) and machine learning (ML) are increasingly proposed as forecasting tools to support early-warning and response. However, the available evidence is fragmented across pathogens, settings and modelling approaches. This review provides, to the best of our knowledge, the first comprehensive comparative assessment of AI/ML models forecasting mosquito-borne viral diseases in human populations, jointly synthesising predictive performance across model families and appraising both methodological quality and operational readiness. Methods: Following PRISMA 2020, we searched PubMed, Embase and Scopus up to August 2025. We included studies applying AI/ML or statistical models to predict arboviral incidence, outbreaks or temporal trends and reporting at least one quantitative performance metric. Given the substantial heterogeneity in outcomes, predictors and time–space scales, we conducted a descriptive synthesis. Risk of bias and applicability were evaluated using PROBAST. Results: Ninety-eight studies met the inclusion criteria, of which 91 focused on dengue. The forecasts spanned national to city-level settings and annual-to-weekly resolutions. Across classification tasks, tree-ensemble models showed the most consistent performance, with accuracies typically above 0.85, while classical ML and deep-learning models showed wider variability. For regression tasks, errors increased with temporal horizon and spatial aggregation: short-term, fine-scale forecasts (e.g., weekly city level) often achieved low absolute errors, whereas long-horizon national models frequently exhibited very large errors and unstable performance. PROBAST assessment indicated that most studies (63/98) were at high risk of bias, with only 24 judged at low risk and limited external validation. Conclusions: AI/ML models, especially tree-ensemble approaches, show strong potential for short-term, fine-scale forecasting, but their reliability drops substantially at broader spatial and temporal scales. Most remain research-stage, with limited external validation and minimal operational deployment. This review clarifies current capabilities and highlights three priorities for real-world use: standardised reporting, rigorous external validation, and context-specific calibration. Full article

The increasing availability of satellite data at different spatial resolutions offers new opportunities for environmental monitoring, highlighting the limitations of medium-resolution products for fine-scale territorial analysis. However, it also raises the need to enhance the resolution of low-quality imagery to enable more detailed spatial assessments. This study investigates the effectiveness of different super-resolution techniques applied to low-resolution (LR) multispectral Sentinel-2 satellite imagery to generate high-resolution (HR) data capable of supporting advanced knowledge extraction. Three main methodologies are compared: traditional bicubic interpolation, a generic Artificial Neural Network (ANN) approach, and a Convolutional Neural Network (CNN) model specifically designed for super-resolution tasks. Model performances are evaluated in terms of their ability to reconstruct fine spatial details, while the implications of these methods for subsequent visualization and environmental analysis are critically discussed. The evaluation protocol relies on RMSE, PSNR, SSIM, and spectral-faithfulness metrics (SAM, ERGAS), showing that the CNN consistently outperforms ANN and bicubic interpolation in reconstructing geometrically coherent structures. The results confirm that super-resolution improves the apparent spatial detail of existing spectral information, thus clarifying both the practical advantages and inherent limitations of learning-based super-resolution in Earth observation workflows. Full article

In machine learning, the Bayes classifier represents the theoretical optimum for minimizing classification errors. Since estimating high-dimensional probability densities is impractical, simplified approximations such as naïve Bayes and k-nearest neighbor are widely used as baseline classifiers. Despite their simplicity, these methods require design choices—such as the distance measures in kNN, or the feature independence in naïve Bayes. In particular, naïve Bayes relies on implicit assumptions by using Gaussian mixtures or univariate kernel density estimators. Such design choices, however, often fail to capture heterogeneous distributional structures across features. We propose a flexible naïve Bayes classifier that leverages Pareto Density Estimation (PDE), a parameter-free, non-parametric approach shown to outperform standard kernel methods in exploratory statistics. PDE avoids prior distributional assumptions and supports interpretability through visualization of class-conditional likelihoods. In addition, we address a recently described pitfall of Bayes’ theorem: the misclassification of observations with low evidence. Building on the concept of plausible Bayes, we introduce a safeguard to handle uncertain cases more reliably. While not aiming to surpass state-of-the-art classifiers, our results show that PDE-flexible naïve Bayes with uncertainty handling provides a robust, scalable, and interpretable baseline that can be applied across diverse data scenarios. Full article

One major factor influencing the development of eco-friendly policies and the implementation of climate change mitigation strategies is the accurate projection of CO₂ emissions. Traditional statistical models face significant limitations in capturing complex nonlinear interactions within high-dimensional emissions data. Advanced deep learning architectures offer new opportunities to overcome these computational challenges due to their strong pattern-recognition capabilities. This paper evaluates four distinct deep learning architectures for CO₂ emissions forecasting: Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRUs), Hybrid Convolutional–LSTM (CNN–LSTM) systems, and Dense Neural Networks (DNNs). A comprehensive comparison is conducted using consistent training protocols, hyperparameters, and performance metrics across five prediction horizons (1, 3, 6, 12, and 24 steps ahead) to reveal architecture-specific degradation patterns. Furthermore, analyzing emissions by category provides insight into the suitability of each architecture for varying levels of pattern complexity. LSTM-based models demonstrate particular strength in modeling long-term temporal dependencies, making them well-suited for integration into long-range environmental policy planning frameworks. Overall, this study provides empirical evidence supporting the use of neural networks in climate modeling and proposes criteria for selecting optimal architectures based on forecasting horizon and computational constraints. Full article

The purpose of the present study is to improve the efficiency of phishing web resource detection through multimodal analysis and using methods of explainable artificial intelligence. We propose a late fusion architecture in which independent specialized models process four modalities and are combined using weighted voting. The first branch uses CatBoost for URL features and metadata; the second uses CNN1D for symbolic-level URL representation; the third uses a Transformer based on a pretrained CodeBERT for the homepage HTML code; and the fourth uses EfficientNet-B7 for page screenshot analysis. SHAP, Grad-CAM, and attention matrices are used to interpret decisions; a local LLM generates a consolidated textual explanation. A prototype system based on a microservice architecture, integrated with the SOC, has been developed. This integration enables streaming processing and reproducible validation. Computational experiments using our own updated dataset and the public MTLP dataset show high performance: F1-scores of up to 0.989 on our own dataset and 0.953 on MTLP; multimodal fusion consistently outperforms single-modal baseline models. The practical significance of this approach for zero-day detection and false positive reduction, through feature alignment across modalities and explainability, is demonstrated. All limitations and operational aspects (data drift, adversarial robustness, LLM latency) of the proposed prototype are presented. We also outline areas for further research. Full article

Machine learning (ML) has become a transformative tool in nanomaterial research, driven by the rapid growth of data-intensive experimental techniques, multiscale simulations, and computational modeling. This study provides a bibliometric analysis to characterize how ML has been integrated into nanomaterial discovery and design. Following a PRISMA-guided workflow, research articles published between 2010 and 2025 were retrieved from Scopus and Web of Science, yielding a curated dataset of 4432 peer-reviewed documents. Here, performance indicators, citation patterns, and network analyses were examined to reveal publication growth, leading journals, productive institutions, and country-level contributions. The results show an exponential increase in scientific output since 2017 and a research landscape dominated by China, the United States, India, and Iran. Keyword co-occurrence and thematic mapping reveal four major research clusters: (i) ML-assisted nanoparticle synthesis, (ii) ML-driven nanocomposite design, (iii) data-driven modeling of carbon-based nanomaterials, and (iv) ML-supported catalysis and nanoscale chemistry. These results demonstrate the rapid consolidation of ML-enabled nanomaterial research and highlight emerging opportunities and challenges. The review provides an integrated summary of the field and highlights key future opportunities for advancing data-driven nanomaterial research. Full article

Fine-grained sentiment analysis requires a deep understanding of emotional intensity in the text to distinguish subtle shifts in polarity, such as moving from positive to more positive or from negative to more negative, and to clearly separate emotionally neutral statements from polarized expressions, especially in short or contextually sparse texts such as social media posts. While recent advances combine deep semantic encoding with context-aware architectures, such as Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Networks (CNNs), many models still struggle to detect nuanced emotional cues, particularly in short texts, due to the limited contextual information, subtle polarity shifts, and overlapping affective expressions, which ultimately hinder performance and reduce a model’s ability to make fine-grained sentiment distinctions. To address this challenge, we propose an Emotion- Aware Bidirectional Gating Network (Electra-BiG-Emo) that improves sentiment classification and subtle sentiment differentiation by learning contextual emotion representations and refining them with auxiliary emotional signals. Our model employs an asymmetric gating mechanism within a BiLSTM to dynamically capture both early and late contextual semantics. The gates are temperature-controlled, enabling adaptive modulation of emotion priors, derived from Reddit post datasets to enhance context-aware emotion representation. These soft emotional signals are reweighted based on context, enabling the model to amplify or suppress emotions in the presence of an ambiguous context. This approach advances fine-grained sentiment understanding by embedding emotional awareness directly into the learning process. Ablation studies confirm the complementary roles of semantic encoding, context modeling, and emotion modulation. Further our approach achieves competitive performance on Sem- Val 2017 Task 4c, Twitter US Airline, and SST5 datasets compared with state-of-the-art methods, particularly excelling in detecting subtle emotional variations and classifying short, semantically sparse texts. Gating and modulation analyses reveal that emotion-aware gating enhances interpretability and reinforces the value of explicit emotion modeling in fine-grained sentiment tasks. Full article

Whole-slide histology images (WSIs) can exceed 100 k × 100 k pixels, making direct pixel-level segmentation infeasible and requiring patch-level classification as a practical alternative for downstream WSI segmentation. However, most approaches either treat patches independently, ignoring spatial and biological context, or rely on deep graph models prone to oversmoothing and loss of local tissue detail. We present WSI-GT (Pseudo-Label Guided Graph Transformer), a simple yet effective architecture that addresses these challenges and enables accurate WSI-level tissue segmentation. WSI-GT combines a lightweight local graph convolution block for neighborhood feature aggregation with a pseudo-label guided attention mechanism that preserves intra-class variability and mitigates oversmoothing. To cope with sparse annotations, we introduce an area-weighted sampling strategy that balances class representation while maintaining tissue topology. WSI-GT achieves a Macro F1 of 0.95 on PATH-DT-MSU WSS2v2, improving by up to 3 percentage points over patch-based CNNs and by about 2 points over strong graph baselines. It further generalizes well to the Placenta benchmark and standard graph node classification datasets, highlighting both clinical relevance and broader applicability. These results position WSI-GT as a practical and scalable solution for graph-based learning on extremely large images and for generating clinically meaningful WSI segmentations. Full article

Transformers for time series forecasting rely on positional encoding to inject temporal order into the permutation-invariant self-attention mechanism. Classical sinusoidal absolute encodings are fixed and purely geometric; learnable absolute encodings often overfit and fail to extrapolate, while relative or advanced schemes can impose substantial computational overhead without being sufficiently tailored to temporal data. This work introduces a family of window-statistics positional encodings that explicitly incorporate local temporal semantics into the representation of each timestamp. The base variant (WinStat) augments inputs with statistics computed over a sliding window; WinStatLag adds explicit lag-difference features; and hybrid variants (WinStatFlex, WinStatTPE, WinStatSPE) learn soft mixtures of window statistics with absolute, learnable, and semantic positional signals, preserving the simplicity of additive encodings while adapting to local structure and informative lags. We evaluate proposed encodings on four heterogeneous benchmarks against state-of-the-art proposals: Electricity Transformer Temperature (hourly variants), Individual Household Electric Power Consumption, New York City Yellow Taxi Trip Records, and a large-scale industrial time series from heavy machinery. All experiments use a controlled Transformer backbone with full self-attention to isolate the effect of positional information. Across datasets, the proposed methods consistently reduce mean squared error and mean absolute error relative to a strong Transformer baseline with sinusoidal positional encoding and state-of-the-art encodings for time series, with WinStatFlex and WinStatTPE emerging as the most effective variants. Ablation studies that randomly shuffle decoder inputs markedly degrade the proposed methods, supporting the conclusion that their gains arise from learned order-aware locality and semantic structure rather than incidental artifacts. A simple and reproducible heuristic for setting the sliding-window length—roughly one quarter to one third of the input sequence length—provides robust performance without the need for exhaustive tuning. Full article

Machine Learning and Knowledge Extraction have evolved from algorithmic tools for pattern recognition into a unifying foundational scientific framework underpinning virtually all of today’s groundbreaking advances, enabling systematic discovery, interpretation and understanding across domains. This paper introduces a comprehensive research agenda that defines currently the future of innovation in Artificial Intelligence. We identify ten interrelated research frontiers that collectively map the transition from data-driven learning to knowledge-centric, trustworthy, and sustainable intelligence. These frontiers span the full spectrum of future AI research: from physics-informed and hybrid architectures that embed causality and domain knowledge, to multimodal and embedded intelligence that ground AI in real-world contexts; from interpretable and responsible design principles that ensure transparency and fairness, to safe and sustainable deployment in open-world environments. Together, these directions delineate a coherent roadmap toward AI systems that not only predict but also explain, reason, and collaborate. Future AI can be seen as a new member of your research lab, an active participant in knowledge creation, driven by interdisciplinary integration, global cooperation, ethical responsibility, and human oversight. By embedding principles of transparency, sustainability, and societal alignment from the outset, we envision AI as both a catalyst for scientific discovery and a cornerstone of responsible technological progress. Full article

The electronic manufacturing industry is relying on automatic and rapid defect inspection of printed circuit boards (PCBs). Two main challenges hinder the accuracy and real-time defect detection: the growing density of electronic component placement and their size reduction, complicating the identification of tiny defects. This systematic review encompasses 56 relevant articles from the Scopus database between 2015 and the first quarter of 2025. This study examines deep learning (DL) architectures and machine learning (ML) algorithms for defect detection in PCB manufacturing. Findings indicate that 78.6% of the articles used models capable of detecting up to six defect types, and 62.5% relied on custom-made datasets. Convolutional neural networks (CNNs) are commonly utilized architectures due to their flexibility and adaptability to a variety of tasks. Still, real-time defect detection remains a challenge because of the complexity and high throughput in production settings. Likewise, accessible datasets are essential for the electronics industry to achieve broad adoption. Hence, architectures capable of learning and optimizing directly in the production line from unlabeled PCB data, without prior training, are necessary. Full article

This paper presents an Artificial Neural Network (ANN) approach for estimating total real and reactive power losses in electrical power systems. Three network architectures were explored: the Multilayer Perceptron (MLP), the Radial Basis Function (RBF) network, and the Generalized Regression Neural Network (GRNN). The main advantage of the proposed methodology lies in its ability to rapidly compute power loss values throughout the system. ANN models are especially effective due to their capacity to capture the nonlinear characteristics of power systems, thus eliminating the need for iterative procedures. The applicability and effectiveness of the approach were evaluated using the IEEE 14-bus test system and compared with the continuation power flow method, which estimates losses using conventional numerical techniques. The results indicate that the ANN-based models performed well, achieving mean squared error (MSE) values below the predefined threshold during both training and validation (0.001). Notably, the networks accurately estimated the total power losses within the expected range, with residuals on the order of 10⁻⁴. Among the models tested, the RBF network showed slightly superior performance in terms of error metrics, requiring fewer centers to meet the established criteria compared to the MLP and GRNN models (11 centers). However, the GRNN achieved the shortest processing time; even so, all three networks produced satisfactory and consistent results, particularly in identifying the critical points of electrical power systems, which is of fundamental importance for ensuring system stability and operational reliability. Full article

Encoder–decoder models are widely used for pixel-level segmentation due to their ability to capture and combine multiscale features. However, skip connections between the encoder and decoder often require cropping to mitigate border pixel loss during convolutions, which can introduce inefficiencies and limit performance. This study explores the potential of modifying these connections by removing direct encoder-to-decoder links to enhance segmentation accuracy. We propose a novel architecture, termed XCC-Net, which features two context-capturing pathways and two symmetric pathways for enlargement. These pathways are interconnected via channels, enabling automated detection of structures with varied shapes. The XCC-Net’s X-shaped architecture links skip connections exclusively between encoder-to-encoder and decoder-to-decoder, omitting direct encoder-to-decoder feature transfers to potentially improve performance. The XCC-Net model was evaluated on multiple medical imaging datasets, including wireless capsule endoscopy (WCE), colonoscopy, and dermoscopy images. Experimental results showed that XCC-Net outperformed state-of-the-art segmentation models, achieving dice coefficients of 91.70%, 89.26%, 87.15%, and 79.07% on the MICCAI 2017 (Red Lesion), PH2, CVC-ClinicDB, and ISIC 2017 datasets, respectively. XCC-Net’s X-shaped architecture, with its unique skip connections, demonstrates improved segmentation performance across various medical imaging tasks. Full article

The increasing sophistication of malware has challenged the effectiveness of conventional detection techniques, motivating the adoption of Graph Neural Networks (GNNs) for their ability to model the structural and semantic information embedded in control flow graphs. While GNNs offer high detection performance, their lack of transparency limits their applicability in security-critical domains. To address this, we present an explainable malware detection framework, which contains a dual explainer. This dual explainer integrates a GNN explainer with a neural subgraph matching approach and the VF2 algorithm. The proposed method identifies and verifies discriminative subgraphs during training, which are later used to explain new predictions through efficient matching. To enhance the generalization of the neural subgraph matcher, we train it using curriculum learning, gradually increasing subgraph complexity to improve matching quality. Experimental evaluations on benchmark datasets demonstrate that the proposed framework retains high classification accuracy while significantly improving interpretability. By unifying explainable graph learning techniques with subgraph matching, the proposed framework enables analysts to gain actionable insights, fostering greater trust in GNN-based malware detectors. Full article

Artificial Intelligence (AI) is increasingly used to enhance project management practices, especially in risk analysis, where traditional tools often lack predictive capabilities. This study introduces an AI-based tool that supports project teams in identifying and interpreting risks through machine learning and integrated documentation features. A synthetic dataset of 5000 project instances was generated using deterministic rules across 27 input variables, enabling the training of multi-output Decision Tree and Random Forest models to predict risk type, impact, probability, and response strategy. Due to the rule-based structure of the dataset, both models achieved near-perfect classification performance, with Random Forest showing slightly better regression accuracy. These results validate the modelling pipeline but should not be interpreted as real-world predictive accuracy. The trained models were deployed within a web platform offering prediction visualization, automated PDF reporting, result storage, and access to a structured risk management plan template. Survey feedback highlights strong user interest in AI-assisted mitigation suggestions, dashboards, notifications, and mobile access. The findings demonstrate the potential of AI to improve proactive risk assessment and decision-making in project environments. Full article

In this study, a structured and methodological evaluation approach for eXplainable Artificial Intelligence (XAI) methods in medical image classification is proposed and implemented using LIME and SHAP explanations for chest X-ray interpretations. The evaluation framework integrates two critical perspectives: predictive model-centered and human-centered evaluations. Predictive model-centered evaluations examine the explanations’ ability to reflect changes in input and output data and the internal model structure. Human-centered evaluations, conducted with 97 medical experts, assess trust, confidence, and agreements with AI’s indicative and contra-indicative reasoning as well as their changes before and after provision of explainability. Key findings of our study include explanation of sensitivity of LIME and SHAP to model changes, their effectiveness in identifying critical features, and SHAP’s significant impact on diagnosis changes. Our results show that both LIME and SHAP negatively affected contra-indicative agreement. Case-based analysis revealed AI explanations reinforce trust and agreement when participant’s initial diagnoses are correct. In these cases, SHAP effectively facilitated correct diagnostic changes. This study establishes a benchmark for future research in XAI for medical image analysis, providing a robust foundation for evaluating and comparing different XAI methods. Full article

A receipt information extraction task requires both textual and spatial analyses. Early receipt analysis systems primarily relied on template matching to extract data from spatially structured documents. However, these methods lack generalizability across various document layouts and require defining the specific spatial characteristics of unseen document sources. The advent of convolutional and recurrent neural networks has led to models that generalize better over unseen document layouts, and more recently, multi-modal transformer-based models, which consider a combination of text, visual, and layout inputs, have led to an even more significant boost in document-understanding capabilities. This work focuses on the joint use of a neural multi-modal transformer and a rule-based model and studies whether this combination achieves higher performance levels than the transformer on its own. A comprehensively annotated dataset, comprising real-world and synthetic receipts, was specifically developed for this study. The open source optical character recognition model DocTR was used to textually scan receipts and, together with an image, provided input to the classifier model. The open-source pre-trained LayoutLMv3 transformer-based model was augmented with a classifier model head, which was trained for classifying textual data into 12 predefined labels, such as date, price, and shop name. The methods implemented in the rule-based model were manually designed and consisted of four types: pattern-matching rules based on regular expressions and logic, database search-based methods for named entities, spatial pattern discovery guided by statistical metrics, and error correcting mechanisms based on confidence scores and local distance metrics. Following hyperparameter tuning of the classifier head and the integration of a rule-based model, the system achieved an overall F1 score of 0.98 in classifying textual data, including line items, from receipts. Full article