Machine Learning and Knowledge Extraction

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. Full article

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. Full article

Large Language Model (LLM) agents depend heavily on multiple external tools such as APIs, databases and computational services to perform complex tasks. However, these tool executions create latency and introduce costs, particularly when agents handle similar queries or workflows. Most current caching methods focus on LLM prompt–response pairs or execution plans and overlook redundancies at the tool level. To address this, we designed a multi-level caching architecture that captures redundancy at both the workflow and tool level. The proposed system integrates four key components: (1) hierarchical caching that operates at both the workflow and tool level to capture coarse and fine-grained redundancies; (2) dependency-aware invalidation using graph-based techniques to maintain consistency when write operations affect cached reads across execution contexts; (3) category-specific time-to-live (TTL) policies tailored to different data types, e.g., weather APIs, user location, database queries and filesystem and computational tasks; and (4) session isolation to ensure multi-tenant cache safety through automatic session scoping. We evaluated the system using synthetic data with 2.25 million queries across ten configurations in fifteen runs. In addition, we conducted four targeted evaluations—write intensity robustness from 4 to 30% writes, personalized memory effects under isolated vs. shared cache modes, workflow-level caching comparison and workload sensitivity across five access distributions—on an additional 2.565 million queries, bringing the total experimental scope to 4.815 million executed queries. The architecture achieved 76.5% caching efficiency, reducing query processing time by 13.3× and lowering estimated costs by 73.3% compared to a no-cache baseline. Multi-tenant testing with fifteen concurrent tenants confirmed robust session isolation and 74.1% efficiency under concurrent workloads. Our evaluation used controlled synthetic workloads following Zipfian distributions, which are commonly used in caching research. While absolute hit rates vary by deployment domain, the architectural principles of hierarchical caching, dependency tracking and session isolation remain broadly applicable. Full article

Ontology learning from unstructured text has become a critical task for knowledge-driven applications in Big Data and Artificial Intelligence. While significant advances have been made in the automatic extraction of concepts and relations using neural and Transformer-based models, the generation of formal Description Logic axioms required for constructing logically consistent and computationally tractable ontologies remains largely underexplored. This paper puts forward a novel pipeline for automated axiom generation through schema mapping. Our paper introduces three key innovations: a deterministic mapping framework that guarantees logical consistency (unlike stochastic Large Language Models); guaranteed formal consistency verified by OWL reasoners (unaddressed by prior statistical methods); and a transparent, scalable bridge from neural extractions to symbolic logic, eliminating manual post-processing. Technically, the pipeline builds upon the outputs of a Transformer-based fusion model for joint concept and relation extraction. We then map lexical relational phrases to formal ontological properties through a lemmatization-based schema alignment step. Entity typing and hierarchical induction are then employed to infer class structures, as well as domain and range constraints. Using RDFLib and structured data processing, we transform the extracted triples into both assertional (ABox) and terminological (TBox) axioms expressed in Description Logic. Experimental evaluation on benchmark datasets (Conll04 and NYT) demonstrates the efficacy of the approach, with expert validation showing high acceptance rates (>95%) and reasoners confirming zero inconsistencies. The pipeline thus establishes a reliable, scalable foundation for automated ontology learning, advancing the field from extraction to formally verifiable knowledge base construction. Full article

Conversational agents are transforming digital interactions across various domains, including healthcare, education, and customer service, thanks to advances in large language models (LLMs). As these systems become more autonomous and ubiquitous, understanding what constitutes high-quality interaction from a user perspective is increasingly critical. Despite growing empirical research, the field lacks a unified framework for defining, measuring, and designing user-perceived interaction quality in human–artificial intelligence (AI) dialogue. Here, we present an integrative review of 125 empirical studies published between 2017 and 2025, spanning text-, voice-, and LLM-powered systems. Our synthesis identifies three consistent layers of user judgment: a pragmatic core (usability, task effectiveness, and conversational competence), a social–affective layer (social presence, warmth, and synchronicity), and an accountability and inclusion layer (transparency, accessibility, and fairness). These insights are formalised into a four-layer interpretive framework—Capacity, Alignment, Levers, and Outcomes—operationalised via a Capacity × Alignment matrix that maps distinct success and failure regimes. It also identifies design levers such as anthropomorphism, role framing, and onboarding strategies. The framework consolidates constructs, positions inclusion and accountability as central to quality, and offers actionable guidance for evaluation and design. This research redefines interaction quality as a dialogic construct, shifting the focus from system performance to co-orchestrated, user-centred dialogue quality. Full article

Artificial intelligence (AI), particularly machine learning and deep learning architectures, has been widely applied to support thyroid cancer diagnosis, but existing evidence on its performance and limitations remains scattered across techniques, tasks, and data types. This systematic review synthesizes recent work on knowledge extraction from heterogeneous imaging and clinical data for thyroid cancer diagnosis and detection published between 2021 and 2025. We searched eight major databases, applied predefined inclusion and exclusion criteria, and assessed study quality using the Newcastle–Ottawa Scale. A total of 150 primary studies were included and analyzed with respect to AI techniques, diagnostic tasks, imaging and non-imaging modalities, model generalization, explainable AI, and recommended future directions. We found that deep learning, particularly convolutional neural networks, U-Net variants, and transformer-based models, dominated recent work, mainly for ultrasound-based benign–malignant classification, nodule detection, and segmentation, while classical machine learning, ensembles, and advanced paradigms remained important in specific structured-data settings. Ultrasound was the primary modality, complemented by cytology, histopathology, cross-sectional imaging, molecular data, and multimodal combinations. Key limitations included diagnostic ambiguity, small and imbalanced datasets, limited external validation, gaps in model generalization, and the use of largely non-interpretable black-box models with only partial use of explainable AI techniques. This review provides a structured, machine learning-oriented evidence map that highlights opportunities for more robust representation learning, workflow-ready automation, and trustworthy AI systems for thyroid oncology. Full article

Prostate cancer is diagnosed through a comprehensive study of histopathology slides, which takes time and requires professional interpretation. To minimize this load, we developed a semi-supervised learning technique that combines transformer-based representation learning and a custom TransUNet classifier. To capture a wide range of morphological structures without manual annotation, our method pretrains DINOv2 on 10,000 unlabeled prostate tissue patches. After receiving the transformer-derived features, a bespoke CNN-based decoder uses residual upsampling and carefully constructed skip connections to merge data from many spatial scales. Expert pathologists identified only 20% of the patches in the whole dataset; the remaining unlabeled samples were contributed by using a consistency-driven learning method that promoted reliable predictions across various augmentations. The model received precision and recall scores of 91.81% and 89.02%, respectively, and an accuracy of 93.78% on an additional test set. These results exceed the performance of a conventional U-Net and a baseline encoder–decoder network. All things considered, the localized CNN (Convolutional Neural Network) decoding and global transformer attention provide a reliable method for prostate cancer classification in situations with little annotated data. Full article

We present a combined ML/NLP (Machine Learning, Natural Language Processing) pipeline for protecting cloud-based APIs (Application Programming Interfaces), which works both at the level of individual HTTP (Hypertext Transfer Protocol) requests and at the access log file reading mode, linking explicitly technical anomalies with business risks. The system processes each event/access log through parallel numerical and textual branches: a set of anomaly detectors trained on traffic engineering characteristics and a hybrid NLP stack that combines rules, TF-IDF (Term Frequency-Inverse Document Frequency), and character-level models trained on enriched security datasets. Their results are integrated using a risk-aware policy that takes into account endpoint type, data sensitivity, exposure, and authentication status, and creates a discrete risk level with human-readable explanations and recommended SOC (Security Operations Center) actions. We implement this design as a containerized microservice pipeline (input, preprocessing, ML, NLP, merging, alerting, and retraining services), orchestrated using Docker Compose and instrumented using OpenSearch Dashboards. Experiments with OWASP-like (Open Worldwide Application Security Project) attack scenarios show a high detection rate for injections, SSRF (Server-Side Request Forgery), Data Exposure, and Business Logic Abuse, while the processing time for each request remains within real-time limits even in sequential testing mode. Thus, the pipeline bridges the gap between ML/NLP research for security and practical API protection channels that can evolve over time through feedback and retraining. Full article

Understanding the decision-making behavior of machine learning models is essential in domains where individual predictions matter, such as medical diagnosis or sports analytics. While explainable artificial intelligence (XAI) methods such as SHAP provide instance-level feature attributions, they mainly summarize typical decision behavior and offer limited support for systematically exploring atypical yet correctly classified cases. In this work, we introduce the Classification Outlier Variability Score (COVAS), a framework designed to support hypothesis generation through the analysis of explanation variability. COVAS operates in the explanation space and builds directly on SHAP value representations. It quantifies how strongly an individual instance’s SHAP-based explanation deviates from class-specific attribution patterns by aggregating standardized SHAP deviations into a single score. Consequently, the applicability of COVAS inherits the model- and data-agnostic properties of SHAP, provided that explanations can be computed for the underlying model and data. We evaluate COVAS on publicly available datasets from the medical and sports domains. The results show that COVAS reveals explanation-space outliers not captured by feature-space outlier detection or prediction uncertainty measures. Robustness analyses demonstrate stability across parameter choices, class imbalance, model initialization, and model classes. Overall, COVAS complements existing XAI techniques by enabling targeted instance-level inspection and facilitating XAI-guided hypothesis formulation. Full article

Age-related macular degeneration (AMD) is a leading cause of irreversible vision loss; anti-vascular endothelial growth factor (anti-VEGF) therapy is standard care for neovascular AMD (nAMD), yet treatment response varies. We systematically reviewed and meta-analyzed artificial intelligence (AI) and machine learning (ML) models using optical coherence tomography (OCT)-derived information to predict anti-VEGF treatment response in nAMD. PubMed, Embase, Web of Science, and IEEE Xplore were searched from inception to 18 December 2025 for eligible studies reporting threshold-based performance. Two reviewers screened studies, extracted data, and assessed risk of bias using PROBAST+AI; pooled sensitivity and specificity were estimated with a bivariate random-effects model. Seven studies met inclusion criteria, and six were synthesized quantitatively. Pooled sensitivity was 0.79 (95% CI 0.68–0.87), and pooled specificity was 0.83 (95% CI 0.62–0.94), with substantial heterogeneity. Specificity tended to be higher for long-term and functional outcomes than for short-term and anatomical outcomes. Most studies had a high risk of bias, mainly due to limited external validation and incomplete reporting. OCT-based AI models may help stratify treatment response in nAMD, but prospective, multicenter validation and standardized outcome definitions are needed before routine use; current evidence shows no consistent advantage of deep learning over engineered radiomic features. Full article

Planning in robotics represents an ongoing research challenge, as it requires the integration of sensing, reasoning, and execution. Although large language models (LLMs) provide a high degree of flexibility in planning, they often introduce hallucinated goals and actions and consequently lack the formal reliability of deterministic methods. In this paper, we address this limitation by proposing a hybrid Sense–Plan–Code–Act (SPCA) framework that combines perception, LLM-based reasoning, and symbolic planning. Within the proposed approach, sensory information is first transformed into a symbolic description of the world in Planning Domain Definition Language (PDDL) using an LLM. A heuristic planner is then used to generate a valid plan, which is subsequently converted to code by a second LLM. The generated code is first validated syntactically through compilation and then semantically in simulation. When errors are detected, local corrections can be applied and the process is repeated as necessary. The proposed method is evaluated in the OpenAI Gym MiniGrid reinforcement learning environment and in a Gazebo simulation on a UR5 robotic arm using a curriculum of tasks with increasing complexity. The system successfully completes approximately 71–75% of tasks across environments with a relatively low number of simulation iterations. Full article

Deep reinforcement learning policies are hard to deploy in safety-critical settings, because they fail to explain why a sequence of actions is taken. We introduce an intrinsically interpretable framework that learns compact summaries of recurring behavior and uses them for case-based decision making. Our method (i) discovers global regimes by grouping trajectories into a small set of recurrent patterns and (ii) learns a prototype-conditioned local policy that maps the current short-horizon pattern to an action (“this matches prototype X → take action Y”). Each action is accompanied by a similarity score to relevant prototypes, which provide the explanations. We evaluate our approach on two domains: (1) CarRacing (pixel-based continuous control) and (2) a real voltage-control problem in low-voltage distribution networks. Our results indicate that the method provides clear pre hoc explanations while keeping task performance close to the reference policy. Full article

In the world of e-commerce, ensuring customer satisfaction and retention depends on delivering an optimal user experience. As the primary point of contact between businesses and consumers, a user interface’s success hinges on personalized human–computer interaction. The goal of this paper is to introduce the concept of a self-adaptive multi-variant user interface based on a novel application of a three-way decision-making model, which allows for “accept”, “reject”, or “delay” decisions on UI changes. The proposed framework enables the delivery of a multi-variant e-commerce user interface. It leverages human-centered machine learning to identify homogeneous groups of customers for whom a layout tailored to their behavior can be offered. The functionality of the solution was verified through pilot implementation and experimental studies. The results positively validated the three-way decision algorithm and highlighted clear directions for its refinement. The primary contribution of this work is the novel adaptation of the three-way decision model to create an automated framework for e-commerce UI personalization, moving beyond the limitations of traditional binary A/B testing. This study demonstrates the practical feasibility of using a self-adaptive, multi-variant interface to significantly improve user experience and key business metrics. These results confirm the feasibility and effectiveness of using self-adaptive e-commerce interfaces to improve the user experience. The proposed framework represents a promising solution to the challenges posed by static interfaces and demonstrates the potential for wider application in the e-commerce domain and beyond. Full article

With the increasing complexity of cyber threats and the inefficiency of traditional vulnerability management, artificial intelligence has been increasingly integrated into cybersecurity. This review provides a comprehensive evaluation of AI-powered strategies including machine learning, deep learning, and large language models for identifying cybersecurity vulnerabilities and supporting automated patching. In this review, we conducted a synthesis and appraisal of 29 peer-reviewed studies published between 2019 and 2024. Our results indicate that AI methods substantially improve the precision of detection, scalability, and response speed compared with human-driven and rule-based approaches. We detail the transition from conventional ML categorization to using deep learning for source code analysis and dynamic network detection. Moreover, we identify advanced mitigation strategies such as AI-powered prioritization, neuro-symbolic AI, deep reinforcement learning and the generative abilities of LLMs which are used for automated patch suggestions. To strengthen methodological rigor, this review followed a registered protocol and PRISMA-based study selection, and it reports reproducible database searches (exact queries and search dates) and transparent screening decisions. We additionally assessed the quality and risk of bias of included studies using criteria tailored to AI-driven vulnerability research (dataset transparency, leakage control, evaluation rigor, reproducibility, and external validation), and we used these quality results to contextualize the synthesis. Our critical evaluation indicates that this area remains at an early stage and is characterized by significant gaps. The absence of standard benchmarks, limited generalizability of the models to various domains, and lack of adversarial testing are the obstacles that prevent adoption of these methods in real-world scenarios. Furthermore, the research suggests that the black-box nature of most models poses a serious problem in terms of trust. Thus, XAI is quite pertinent in this context. This paper serves as a thorough guide for the evolution of AI-driven vulnerability management and indicates that next-generation AI systems should not only be more accurate but also transparent, robust, and generalizable. Full article

Environmental monitoring systems generate large volumes of multivariate time series data from heterogeneous sensors, including those measuring soil, weather, and air quality parameters. However, sensor malfunctions and transmission failures frequently lead to missing values, compromising the performance of downstream analytical and predictive models. To address this challenge, this study presents a comprehensive and systematic evaluation of previously proposed hybrid architecture that interleaves Long Short-Term Memory (LSTM) layers with a Multi-Head Attention mechanism in a “sandwiched” setting (LSTM–Attention–LSTM) for robust multivariate data imputation in environmental IoT datasets. The first LSTM layer captures short-term temporal dependencies, the attention layer emphasises long-range relationships among correlated features, and the second LSTM layer re-integrates these enriched representations into a coherent temporal sequence. The model is evaluated using multiple environmental datasets of soil temperature, meteorological (precipitation, temperature, wind speed, humidity), and air quality data across missingness levels ranging from 10% to 90%. Performance is compared against baseline methods, including K-Nearest Neighbour (KNN) and Bidirectional Recurrent Imputation for Time Series (BRITS). Across all datasets, the Hybrid model consistently outperforms baseline methods, achieving MAE reductions exceeding 50% and reaching over 80% in several scenarios, along with RMSE reductions of up to approximately 85%, particularly under moderate to high missingness conditions. An ablation study further examines the contribution of each layer to overall model performance. Results demonstrate that the proposed Hybrid model achieves superior accuracy and robustness across datasets, confirming its effectiveness for environmental sensor data imputation under varying missing data conditions. Full article

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