Foundations of Computing and Decision Sciences Feed

SkinNet: Explainable Lightweight CNN Model for Skin Disease Classification

Sun, 11 Oct 2026 00:00:00 GMT

Dermatological disorders constitute a significant global health burden, yet the accurate and timely diagnosis remains a challenge due to the subjective nature of visual inspections and the lack of specialists. Although deep learning (DL) has shown promise in automated classification, the black box nature of these models and their high computational costs often hinder clinical adoption. This study proposes a lightweight, interpretable Custom Convolutional Neural Network (CNN) for the multi-class classification of five distinct skin conditions: Acne, Vitiligo, Hyperpigmentation, Nail Psoriasis, and Stevens-Johnson syndrome/toxic epidermal necrolysis (SJS-TEN). Using a data set of 9,548 images, we implemented a comprehensive pipeline featuring a two-stage augmentation and balance strategy to mitigate class imbalance. The new proposed SkinNet model was benchmarked against four established architectures: ZooCNN, CNN, MobileNetV2, and DenseNet121. Experimental results demonstrate that SkinNet achieved superior performance with an overall accuracy of 86% and a macro-averaged F1-score of 0.85, outperforming transfer learning baselines. To address the critical need for transparency in medical AI, the model’s decision-making process was validated using four Explainable AI (XAI) techniques: Local Interpretable Model-agnostic Explanations (LIME), Gradient-weighted Class Activation Mapping (Grad-CAM), Integrated Gradients, and Diverse Counterfactual Explanations (DiCE). These visualizations confirmed that the model consistently focuses on relevant pathological features such as inflammatory papules, depigmented patches, and onycholysis rather than background noise. This research highlights the efficacy of task-specific, lightweight architectures to provide accurate, efficient, and reliable diagnostic support for dermatology

Intelligent Multi-Stage Feature Selection for Real-Time Intrusion Detection in IoV Networks: A Machine Learning Framework

Sun, 11 Oct 2026 00:00:00 GMT

The Internet of Vehicles (IoV) faces critical cybersecurity challenges that require real-time intrusion detection on resource-constrained edge devices. This study proposes a Three-stage Variance-correlation Binary Grey Wolf Optimization (TVC-BGWO) feature selection framework integrated with ensemble learning. The framework combines Variance Threshold filtering, Correlation-based filtering, and Binary Grey Wolf Optimization to achieve 93.59% feature reduction (78→5 features) while maintaining competitive detection performance. Random Forest with TVC-BGWO achieved 99.91% binary accuracy and 99.87% multiclass accuracy 6 attack types plus benign traffic on Canadian Institute for Cybersecurity Intrusion Detection System (CIC-IDS2017). Statistical validation (McNemar’s χ2 = 82.01, p < 0.001) confirms a significant improvement. The framework provides 59.0% training time reduction and sub-millisecond inference latency meeting V2V communication requirements. Adversarial evaluation reveals exceptional resilience to data poisoning 99.98% robustness and moderate vulnerability to test-time perturbations. This work contributes to intelligent transportation security by demonstrating competitive intrusion detection performance while achieving substantial computational efficiency gains suitable for edge deployment.

Exploring Tunisian Primary School Teachers’ Perceptions of ChatGPT Integration in Education: Insights from the Technology Acceptance Model (TAM)

Sun, 11 Oct 2026 00:00:00 GMT

This study investigates the acceptance of ChatGPT as an educational tool through the lens of the Technology Acceptance Model (TAM), a widely recognized theoretical framework for understanding technology adoption. The primary objective is to examine how primary school teachers perceive ChatGPT in terms of its usefulness and ease of use within educational settings. The TAM framework emphasizes two core dimensions: Perceived Usefulness (PU) and Perceived Ease of Use (PEOU). Additional factors, including attitude, behavioral intention, perceived credibility, and social influence, were incorporated to provide a comprehensive analysis.A survey was conducted among a sample of teachers to capture their perceptions regarding the integration of ChatGPT into teaching practices. The findings indicate that both PU and PEOU significantly influence teachers’ attitudes and their intention to use the tool. This research highlights the pedagogical potential of ChatGPT and offers practical recommendations for its effective integration into classroom instruction. By applying the TAM framework, the study contributes to a deeper understanding of the factors shaping the adoption of generative AI in education.

An Accelerated L-shaped Method for Optimizing Over the Stochastic Efficient Set

Sun, 11 Oct 2026 00:00:00 GMT

In this paper, we propose a new exact method to optimize a stochastic linear function over an efficient set of a multi-objective stochastic integer linear programming (MOSILP) problem. The proposed method combines two approaches: the first is called the L method, while the second is an adaptation of the branch-andbound procedure. This adaptation is reinforced by efficient cuts and tests, enabling the method to eliminate many inefficient solutions in the search tree. A didactic example is provided to detail all steps of the method. This is followed by testing the method on a set of randomly generated instances and comparing the obtained results with the most recent method. The experimental study shows that the proposed method remains competitive and delivers satisfactory results.

On the Problem of Latent Temporal Modulation in Diagnostic Decision-Making

Fri, 26 Jun 2026 00:00:00 GMT

Modern diagnostic systems increasingly rely on longitudinal and multimodal data, yet they typically assume a stable relation between observations and disease risk. We introduce latent temporal modulation (LTM) as a conceptual framework for a class of problems in which the relation (i.e., the mapping between observations and risk) evolves over time due to temporally structured, non-observable factors. We formalise LTM and show that it is not explicitly represented within standard dynamic or latent-state modelling frameworks. In such settings, diagnostically relevant information is encoded in the evolving relation between observations and disease risk rather than in observed variables alone. This perspective has implications for the design of diagnostic decision systems operating on temporal medical data. In particular, LTM shifts the focus of diagnostic modelling from analysing temporal data alone to analysing how their interpretation evolves over time.

Investigation and RL-based Optimization of Offloading Decision Making in Industrial Edge Computing Scenarios

Fri, 26 Jun 2026 00:00:00 GMT

Computation Offloading (CO) transfers computationally intensive tasks from hardware-limited devices (clients) to more powerful servers. This enables clients to overcome performance and power limitations and meet Quality of Service requirements. Research in CO focuses on “Why,” “What,” “If/When,” “Where,” and “How” to offload. This work investigates the “If/When” question from a performance perspective in dynamic environments. To optimize decision making, a Reinforcement Learning (RL) agent was designed, implemented, and trained to make efficient offloading decisions. The practicality of our solution was tested by emulating real hardware in industrial environments using 5G New Radio. Experimental results show that the proposed RL agent outperforms the existing closed-form solution from the literature, improves the total completion time of emulated apps, and reduces client device power consumption.

A Hybrid Framework Integrating Novel Segmentation and Hybrid Feature Engineering with M-SVM for the Detection and Classification of Wheat Rusts and Septoria

Fri, 26 Jun 2026 00:00:00 GMT

This study introduces a hybrid framework for the precise detection and classification of wheat rusts (Leaf Rust, Stem Rust, and Stripe Rust) and Wheat Leaf Septoria, aiming to enhance diagnostic accuracy and support effective disease management. The framework comprises key stages: lesion segmentation, Hybrid Feature Engineering and disease classification. Advanced machine learning and image enhancement methods are combined to effectively segment lesions using a multilevel hybrid segmentation model, followed by feature extraction involving shape, color, texture, and edge properties refined through PCA, entropy and skewness. To address challenges related to limited datasets and model overfitting, robust handcrafted features are utilized instead of deep feature dependency. Finally, a Multi-Class Support Vector Machine (M-SVM) is employed for classification.Evaluated on four wheat diseases—septoria, stem rust, leaf rust, and stripe rust—the framework achieved a segmentation performance index of 98.21% on 40 images per disease and a classification accuracy of 99.1% using 10-fold cross-validation on 1,000 images. With an F1-score of 0.989, AUC of 0.991, and minimal error rates, the system outperforms existing methods, averaging 5.8 seconds per image, demonstrating its robustness and efficiency.

The Labeling Dilemma: Study of Clickbait Datasets and their Methodologies

Fri, 26 Jun 2026 00:00:00 GMT

In today’s digital ecosystem, there is a dominance of attention driven content platforms that promote sensationalism over informational quality. These platforms use various means to manipulate users. Clickbait is among them. It often uses misleading or exaggerated headlines to lure people to click on a link. This leaves them in discontent as the promises are never met. The aim is to gain user engagement by either routing them to a page with lots of advertisements that, in turn, boost their revenue or simply spreading misinformation. This necessitate the development of automatic clickbait detection models. This article serves as a systematic review of the work done in this domain, focusing on two important areas: existing datasets with their labeling techniques, and the evolution of various clickbait detection models from ML to DL to new pre-trained language models such as BERT and RoBERTa. This paper aims to serve academic researchers and industry professionals seeking an overview of clickbait detection methods with particular emphasis on ground truth datasets generation and their labeling strategies.

An Enhanced Intuitionistic Fuzzy TOPSIS approach Based on Yager’s Aggregation Operators and the Application in Selecting Green Suppliers

Fri, 26 Jun 2026 00:00:00 GMT

The theory of intuitionistic fuzzy sets (IFS) is ideally suited to handle uncertainty and haziness. In this study, an enhanced fuzzy TOPSIS-based method for dealing with multi-attribute group decision making (MAGDM) problems under intuitionistic fuzzy information, where the weights of the decision-makers (DMs) and criteria are completely unknown. Firstly, the Yager operational rules are initiated for intuitionistic fuzzy numbers (IFNs) constructed on Yager T-norm (TN) and T-conorm and various core properties of these operational rules are investigated. Secondly, utilizing these operational various weighted aggregation operators such as, intuitionistic fuzzy Yager weighted averaging (IFYWA) operator, intuitionistic fuzzy Yager weighted ordered weighted averaging (IFYOWA) operator, intuitionistic fuzzy Yager weighted hybrid averaging (IFYWHA) operator, intuitionistic fuzzy Yager weighted geometric (IFYWG) operator, intuitionistic fuzzy Yager ordered weighted geometric (IFOWG) operator, intuitionistic fuzzy Yager hybrid weighted geometric (IFYHWG) operator are initiated. Thirdly, a few characteristics of the intended aggregation operators are investigated. Fourthly, a novel MAGDM model is constructed based on the intended aggregation operators to handle IF information. Finally, a numerical example related to the selection of suitable green suppliers is provided to show the efficacy and practicality of the initiated MAGDM approach, along with the comparison to some existing MAGDM approaches.

A Deep Learning Approach to Classifying Software Requirements: The Application of Transformer-Based Ensemble Learning and Attention-Based Fusion

Tue, 17 Mar 2026 00:00:00 GMT

The functional requirements (FRs) classification in software requirements classification (SRC) is a difficult task due to class imbalance, fine-grained subcategories, and semantic complexities. Existing Machine Learning (ML) and Deep Learning (DL) models often rely on handcrafted characteristics or overlook the contextual meaning. This work presents a novel hybrid ensemble framework that refines three pre-trained transformers (BERT, DistilBERT, and RoBERTa) and combines them using two mechanisms: (1) an Attention-Based Fusion Mechanism that dynamically selects the most contextually relevant transformer for each instance, and (2) an Accuracy-Per-Class Weighted Ensemble that assigns weights based on per-class validation accuracy. Tested on multiple datasets, the approach outperformed single-transformer and DL models (CNN, LSTM, BiLSTM, and GRU) by a large margin (p < 0.001), achieving 95% accuracy and 0.94 F1-score. To the best of our knowledge, this is the first study to combine attention fusion and transformer-based ensembles for SRC, establishing a new standard for SRC.

Machine Learning for Multi Objective Convex Separable Programming (MOCSP) with Aggregation of Linear Approximations and Portfolio Optimization

Tue, 17 Mar 2026 00:00:00 GMT

A novel technique is developed for nonlinear optimization problem which is convex, separable and having multiple objective functions. In the development of the model all the objectives and the constraints of the multi objective model are linearly approximated over suitable intervals. The linear approximations are then aggregated to account for the original problem. The developed technique has been utilized for portfolio optimization problem. Firstly, the minimum variance model has been formulated and solved with machine leaning techniques. Secondly, the risk aversion model has been formulated and solved. The results obtained are combined into a multi objective framework of convex separable programming problem. All the three problems have been solved with the help of the XGBoost, neural network, and decision forest regression models. The renowned Python machine libraries of scikit-learn and keras have been utilized. The results identified portfolios that can return more financial benefits to the investors while investing in the capital market. The results of the proposed MOCSP approach are 22.5% improved in case of risk aversion model. Additionally, 17% improvement has been recorded in case of the minimum risk model. The MAE and RMSE for both XGBoost and decision forest regression have a frail value 0.0001. MAE and RMSE for the neural network regression have been recorded 1% and 2%, respectively. Both Accuracy and F1 score for XGBoost are 91%, for neural network regression are 98%, and for decision forest are 92%, respectively.

MOPOA: A New Multi-Objective Pufferfish Optimization Algorithm

Tue, 17 Mar 2026 00:00:00 GMT

Multi-objective optimization problems (MOPs) pose significant challenges due to the presence of multiple conflicting objectives. This paper introduces MOPOA, a novel Multi-Objective Pufferfish Optimization Algorithm inspired by the defensive behaviors of pufferfish in nature. MOPOA extends the original single-objective POA by incorporating Pareto dominance, an external archive for preserving non-dominated solutions, and a crowding distance mechanism to maintain solution diversity. The algorithm balances exploration and exploitation through biologically inspired phases simulating predator-prey interactions. To evaluate MOPOA’s performance, it was benchmarked against several state-of-the-art algorithms, including NSGA-III, MOPSO, MODA, and MOFDO, on two well-known test suites: the ZDT and CEC-2019 multi-objective functions. Results indicate that MOPOA not only achieves superior convergence to the Pareto front but also maintains high diversity and robustness across diverse optimization scenarios. These findings position MOPOA as a powerful and adaptive tool for solving complex real-world multi-objective problems.

The Potential of Prospect Theory under Inductive Reasoning

Tue, 17 Mar 2026 00:00:00 GMT

The TODIM method was the first Multiple Criteria Decision Making (MCDM) approach to incorporate the advances of judgment and decision theory based on the findings of prospect theory. Since its advent, several studies have proposed new methods for the same purpose, such as Behavioral TOPSIS. Additionally, new versions of the original TODIM method have been developed to enhance its adherence to prospect theory. Among these approaches, the ExpTODIM version demonstrated high accuracy in predictions under deductive reasoning scenarios. In this research, an inductive reasoning analysis of the potential of ExpTODIM and Behavioral TOPSIS is conducted. The results are then compared with a native inductive reasoning method, namely multinomial regression analysis. It was found that ExpTODIM demonstrated strong potential for applications under an inductive reasoning perspective, which supports the application of prospect theory in tasks such as autonomous classification, a key area in artificial intelligence studies.

Influence of Graphical Representation Type on Tessellated Geometry Classification

Tue, 17 Mar 2026 00:00:00 GMT

This research investigates the impact of graphical representation on the accuracy of three-dimensional shape classification. 3D models, both scanned and modeled, are used in engineering, computer graphics and scientific data visualization. Various approaches are adopted in these fields to visually represent 3D geometry, utilizing solutions such as OpenGL and Direct3D, each with its distinct goal of achieving either real-time manipulation or photorealism. The purpose of this research was to determine the most effective graphical representation for categorizing mechanical components with a high degree of geometric similarity, such as beams and rods. The study examined various image representations and their combinations, obtained through the adjustment of rendering parameters and image compositing. In an effort to improve classification accuracy, novel techniques for addressing image recognition issues were developed and tested against commonly used image representation methods. This innovative approach proposed in the paper led to a 48% reduction in classification errors.

A Transparent AI-Driven Multiclass Decision Support System for Thyroid Risk Prediction Using Machine Learning and Deep Learning Approaches

Mon, 08 Dec 2025 00:00:00 GMT

Early and accurate diagnosis of thyroid disorders is essential due to their prevalence and health impact. To enhance interpretability in clinical settings, we propose a comprehensive workflow for transparent thyroid disease prediction using a multiclass classification problem with five diagnostic categories. A dataset of 9172 samples with 31 features was used to train various machine and deep learning models. A dual-layered framework combining Feature Selection (ETC, MI, RFE) and Explainable AI (SHAP, LIME) improved performance and transparency. Gradient Boosting achieved the highest accuracy (0.97). SHAP explained global feature influence, while LIME clarified individual predictions. Our approach supports interpretable, reliable AI-based diagnostic tools for thyroid disorder classification.

DermoAI-CNN: Leveraging GANs, Mask R-CNN, and Attention Mechanisms for Enhanced Skin Disease Analysis

Mon, 08 Dec 2025 00:00:00 GMT

The study aims to develop an effective and efficient deep learning model for detecting skin diseases, as skin diseases rank as the world’s number one health problem. Besides, cancers and dermatological anomalies should be diagnosed at an early stage, so that subsequent treatment can be efficient and complication-free. The existing methods of diagnosis are associated with lower precision and, in most cases, are inefficient, which can be attributed to the lack of effective data augmentation, segmentation techniques, and improved feature extraction. In this paper, a general framework is introduced that uses Generative Adversarial Networks for data augmentation, Mask R-CNN for precise segmentation, and a tailored multilayer Convolutional Neural Network with an attention mechanism incorporated into it to classify 23 skin disease classes using 25,250 images, among them 5,750 generated by GAN, to balance underrepresented classes. The accuracy attained was 97.30%, which was much better than that reported in earlier studies, which ranged from 85 to 92. The metrics, including an accuracy of 95.65%, a recall of 97.09%, and an F1-score of 96.98%, were used to assess the system’s performance in classifying invisible dermatological images. The scalable system provides explanations that support real-time diagnosis, preventing delays and acute health costs. The findings fully fulfil the capabilities of deep learning in dermatology, as the initial diagnosis of the skin disease is accurate, accessible and efficient.

Quantifying the Digital Divide: A Data-Driven Approach with EFA, Shannon Entropy and Sensitivity Testing

Mon, 08 Dec 2025 00:00:00 GMT

In the digital world, inequalities in digital participation often arise as a consequence of the uneven distribution of digital resources. Therefore, the objective of this study is to develop a digital divide index (DDI) that facilitates the temporal and spatial examination of disparities in the availability and quality of internet access, internet use and outcomes. Empirical data for 27 European countries were collected from online available datasets for 2014 and 2022. Exploratory factor analysis (EFA) was used to identify factors in DDI, while the weight values were calculated using the Shannon entropy method. The average difference in DDI scores indicates a spread of the digital divide facilitated by Finland, Latvia, Estonia, Germany, Poland and Slovakia. States in Northern and Central Europe like Norway, Austria and the Czech Republic are at the forefront of achieving digital equality. Spatial analysis reveals that Benelux countries along with Scandinavian countries show the highest levels of DDI, while Southern Europe lags behind. Sensitivity testing results show a stable index structure with variations in the importance of factors not significantly affecting the ranking results.

f-divergence Analysis of Generative Adversarial Network

Mon, 08 Dec 2025 00:00:00 GMT

We aim to establish estimation bounds for various divergences, including total variation, Kullback-Leibler (KL) divergence, Hellinger divergence, and Pearson χ2 divergence, within the GAN estimator. We derive an inequality based on empirical and population objective functions of the GAN model, achieving almost surely convergence rates. Subsequently, this inequality was employed to derive estimation bounds for total variation, Kullback-Leibler (KL) divergence, Hellinger divergence, and Pearson χ2 divergence, leading to almost surely convergence rates and differences between the expected outputs of the discriminator on real data and generated data. Our study demonstrates better results compared to some existing ones, which are a specific case of the general objective function.

Leveraging Artificial Intelligence for Cyanobacterial Bloom Prediction: A Hybrid Deep Learning and Generative Adversarial Network Framework for Accurate Forecasting and Proactive Management

Mon, 08 Dec 2025 00:00:00 GMT

This study presents an Artificial Intelligence-based system designed to predict cyanobacterial harmful algal blooms (CyanoHABs). The system utilizes Long Short-Term Memory (LSTM) networks to predict the timing of bloom occurrences and One-Dimensional Convolutional Neural Networks (1D-CNNs) to estimate cyanobacterial density. Additionally, Generative Adversarial Networks (GANs) are employed for data augmentation to enrich the database. The system’s performance was validated using the Algerian Mexa database, achieving an R-squared (R2) value of 98% and a root mean square error (RMSE) of 9% for cyanobacterial density prediction, and an R-squared value of 88% with a root mean square error of 31% for bloom timing prediction. These results highlight the system’s robust predictive capabilities, enabling proactive monitoring and management of CyanoHABs to mitigate their adverse impacts on health and the environment.

Preface to the Issue on Recent Developments in AI-Based Robotics: A Polish Perspective

Thu, 21 Aug 2025 00:00:00 GMT