IJACSA Vol. 17 Issue 2 (2026)

1

A Comparison of Metaheuristic Methods for the Vehicle Routing Problem

Author 1: Manal El Jaouhari Author 2: Ghita Bencheikh Author 3: Ghizlane Bencheikh

The Capacitated Vehicle Routing Problem (CVRP) is a fundamental NP-hard combinatorial optimization problem with important applications in logistics and distribution systems. Although numerous advanced approaches have been proposed in recent years, systematic benchmarking of classical metaheuristic algorithms under a unified experimental framework remains limited. This study evaluates the performance and… Read full abstract & cite →

Metaheuristics Ant Colony Optimization Hill Climbing Genetic Algorithm Particle Swarm Optimization exact algorithm Capacitated Vehicle Routing Problem

Abstract doi.org/10.14569/IJACSA.2026.0170201 PDF

2

A Confidence-Aware Multi-Layer Framework for Drone Forensic Inference Using Heterogeneous Digital Evidence Sources and Flight Logs Analysis

Author 1: Nidhiba Parmar Author 2: Naveen Kumar Chaudhary

The increasing use of unmanned aerial vehicles (UAVs) in criminal and adversarial contexts has created new challenges for digital forensic investigations. Current UAV forensic research primarily emphasizes artefact extraction and platform-dependent analysis, while insufficient attention has been given to uncertainty modelling and confidence quantification in forensic inference. This study addresses… Read full abstract & cite →

UAV forensics drone investigations forensic inference confidence quantification flight log analysis digital forensic framework

Abstract doi.org/10.14569/IJACSA.2026.0170202 PDF

3

Real-Time Data-Driven Decision Support in Retail: A Hybrid GraphSAGE+XGBoost Model for Predicting Reorder Behavior and Unraveling Consumer Communities

Author 1: Balayet Hossain Author 2: Md Deluar Hossen Author 3: Md Nuruzzaman Pranto Author 4: Belal Hossain Author 5: Sabrina Shamim Moushi Author 6: Nusrat Ameri Author 7: Khandakar Rabbi Ahmed

The rising demand for real-time, data-driven decision support in retail platforms has underscored the need for intelligent systems capable of modeling both behavioral sequences and product relationships. This study introduces a hybrid architecture for real-time decision support in retailing by coupling graph-based learning with conventional machine learning methods. Based on… Read full abstract & cite →

Real-time business intelligence graph analytics machine learning GraphSAGE XGBoost retail analytics recommendation systems decision support instacart dataset hybrid model

Abstract doi.org/10.14569/IJACSA.2026.0170203 PDF

4

Evaluating the Efficiency of LLM-Generated Software in Resisting Malicious Attacks

Author 1: Dominic Niceforo Author 2: Haydar Cukurtepe

This study introduces a structured framework for evaluating the security of Java applications generated by large language models (LLMs) and presents the results from its implementation across three models: DeepSeek, GPT-4, and Llama 4. The framework integrates Open Web Application Security Project (OWASP)-supported tools, such as SpotBugs with FindSecBugs, OWASP… Read full abstract & cite →

Artificial intelligence cyber security large language models (LLMs) software security evaluation

Abstract doi.org/10.14569/IJACSA.2026.0170204 PDF

5

Beyond the Interface: AI Integration Through Input Stream Mediation and Intelligent Output Simulation

Author 1: Divij H. Patel

As Natural Language Processing (NLP) technologies continue to advance, their integration within everyday software tools remains fragmented and often constrained by environment-specific plugins or proprietary AI interfaces. This study introduces a lightweight, platform-independent framework that transforms any text-input surface, ranging from simple editors to browsers and full-featured integrated development environments… Read full abstract & cite →

Keystroke dynamics keystroke simulation AI-Assisted Interfaces open-source AI key-chord recognition API integration real-time interaction systems

Abstract doi.org/10.14569/IJACSA.2026.0170205 PDF

6

Best Practices to Train Accurate Deep Learning Models: A General Methodology

Author 1: Alberto Nogales Author 2: Ana M. Maitín Author 3: Álvaro J. García-Tejedor

In recent years, the field of computer science has experienced great changes due to the incredible advances in the field of artificial intelligence. Deep Learning models are responsible for most of them since the biggest milestones occurred in 2012 when AlexNet won the image classification challenge called ImageNet. These models… Read full abstract & cite →

Methodology artificial intelligence deep learning

Abstract doi.org/10.14569/IJACSA.2026.0170206 PDF

7

A Constraint-Driven Conversational Architecture for Staged Task Interaction Using Large Language Models

Author 1: Joseph Benjamin Ilagan Author 2: Jose Ramon Ilagan

Large language models (LLMs) enable flexible conversational interfaces, but remain difficult to deploy in structured, staged tasks that require controlled progression, bounded information disclosure, and task validity. Prompt-based control is inherently probabilistic and has been shown to degrade under multi-turn interaction, leading to premature solution disclosure, stage skipping, loss of… Read full abstract & cite →

Large language models conversational systems task-oriented dialogue deterministic control constraint enforcement staged interaction

Abstract doi.org/10.14569/IJACSA.2026.0170207 PDF

8

Measurement of Dataset Quality and Computational Quality of Graph Neural Networks on Analog Integrated Circuit Recognition System

Author 1: Arif Abdul Mannan Author 2: Koichi Tanno

To meet the needs of AI design in analog IC design automation and especially in big data applications, the application of sensitive and precise Graph Neural Network (GNN) recognition will become increasingly necessary. This sensitive GNN recognition is needed to accommodate the highly stringent trade-offs on analog IC design requirements… Read full abstract & cite →

Big data Graph Neural Network Artificial Intelligence calculation quality measurement analog circuit design

Abstract doi.org/10.14569/IJACSA.2026.0170208 PDF

9

LayCoder: UI Layout Completion with an Encoder-Only Transformer and Layout Tokenizer

Author 1: Iskandar Salama Author 2: Luiz Henrique Mormille Author 3: Masayasu Atsumi

The growing complexity of user interface (UI) design calls for effective methods to understand, complete, and refine layout structures. While prior work has focused pre-dominantly on generating UI layouts from scratch, completing partially designed interfaces is equally critical, particularly in iterative design workflows and scenarios involving incomplete prototypes. In this… Read full abstract & cite →

Layout completion deep learning encoder-only transformers masked language modeling tokenization

Abstract doi.org/10.14569/IJACSA.2026.0170209 PDF

10

Reducing Interference in Human–Robot Collaboration: A Distance-Based Policy for Disassembly Collaborative Task

Author 1: Wendy Cahya Kurniawan Author 2: Wen Liang Yeoh Author 3: Osamu Fukuda

This study investigates human-robot collaborative (HRC) disassembly tasks, focusing on the effects of constrained and unconstrained safety distances on human behavior and movement interference. Separate and shared target configurations were compared using a screw-picking task as a prototype for electronic waste (e-waste) disassembly. The system employ a vision-based approach to… Read full abstract & cite →

Human-robot collaboration disassembly collaborative task UI feedback distance policy vision-based sensor

Abstract doi.org/10.14569/IJACSA.2026.0170210 PDF

11

Understanding Authentication and Authorization: A Comparative Analysis of Role-Based Access Control (RBAC), Attribute-Based Access Control (ABAC), and Relationship-Based Access Control (ReBAC) Authorization Models

Author 1: Madhuri Margam

This study elucidates the distinctions between authentication and authorization within information security, two fundamental yet frequently conflated concepts. While authentication serves to confirm an entity’s identity, authorization determines the permissible actions that entity may execute. A thorough understanding of these mechanisms is critical for architecting secure, scalable systems and reducing… Read full abstract & cite →

Authentication authorization access control Role-Based Access Control (RBAC) Attribute-Based Access Control (ABAC) Relationship-Based Access Control (ReBAC) information security least privilege Zero Trust

Abstract doi.org/10.14569/IJACSA.2026.0170211 PDF

12

Construction of an International Trade Financial Risk Assessment and Prediction Model Based on Big Data Analysis

Author 1: Zeyu Liu

Background: International trade promotes economic growth across nations, while imposing financial risks of currency fluctuations, credit defaults, and market volatility. Although conventional methods of risk evaluation have served well in the past, they are, however, unable to provide risky international trade answers under the dynamic conditions at present. Objective: This… Read full abstract & cite →

International trade financial risk big data predictive modeling China’s economy

Abstract doi.org/10.14569/IJACSA.2026.0170212 PDF

13

Attention and Representation Learning in Byte-Level Digital Forensics: A Survey of Methods, Challenges, and Applications

Author 1: Teena Mary Author 2: Sreeja CS

Byte-level analysis has become an essential capability in digital forensics, enabling content-based investigation when file system metadata, headers, or structural information are unavailable or unreliable. Recent advances in deep learning allow forensic systems to learn discriminative features directly from raw byte streams; however, the growing diversity of representation strategies, architectural… Read full abstract & cite →

Byte-level digital forensics representation learning attention mechanisms file fragment classification deep learning forensic robustness

Abstract doi.org/10.14569/IJACSA.2026.0170213 PDF

14

Estimation of Landslide Hazard Zones Using Deep Learning Based on Diverse Geospatial Data

Author 1: Kohei Arai Author 2: Kengo Oiwane Author 3: Hiroshi Okumura

Traditional landslide hazard mapping in Japan relies on labor-intensive field surveys, which are slow, costly, and fail to update dynamically amid rising climate-driven disasters like the 2018 Heavy Rain Event, leaving gaps in timely evacuations. This study addresses these challenges by proposing a semantic segmentation framework using ResUNet to fuse… Read full abstract & cite →

SAR Optical ResUNet++ land cover classification Digital Elevation Model landslide hazard zones The Heavy Rain Event of July 2018

Abstract doi.org/10.14569/IJACSA.2026.0170214 PDF

15

A Hybrid Framework Integrating GNN-LSTM-CNN to Map the Impact of MSME User Behavior on Digital Transformation

Author 1: Jani Kusanti Author 2: Erni Widiastuti Author 3: Bintara Sura Priambada Author 4: Ramadhian Agus Triono Sudalyo Author 5: Masdava Aviv Masyayissa Author 6: Rizki Adhi Pratama

MSMEs need a recommendation system that simultaneously captures the evolution of user intent over time and the relationship structure between entities (users, products, sessions, categories, and security events). The problem with this research is that LSTM excels in sequence, but its performance drops on a rare timeline, a general situation… Read full abstract & cite →

CNN e-commerce hybrid model LSTM MSME user behavior prediction

Abstract doi.org/10.14569/IJACSA.2026.0170215 PDF

16

A Computer Vision–Based Method for Determining the Vaccine Injection Position in Pangasius Fingerlings

Author 1: Nguyen Phuc Truong Author 2: Luong Vinh Quoc Danh Author 3: Nguyen Chanh Nghiem

Pangasius farming in the Mekong Delta is a major component of Vietnam’s aquaculture industry, characterized by large-scale production, intensive farming practices, and significant contributions to export revenue. However, vaccination of Pangasius fingerlings is still predominantly performed manually, resulting in low productivity, high labor demand, and inconsistent injection accuracy, which limit… Read full abstract & cite →

Computer vision fish length measurement fish vaccination injection position OpenCV Pangasius

Abstract doi.org/10.14569/IJACSA.2026.0170216 PDF

17

Exposure-Based Media Mix Modeling Using Machine Learning and Genetic Algorithms

Author 1: Thejan Dulara Author 2: Indra Mahakalanda Author 3: Prasanga Jayathunga

Media budget allocation remains a persistent challenge in the advertising industry. Inefficient spending and biased planning decisions often reduce campaign effectiveness. Advertisers struggle to balance investments across television, radio, press, and digital platforms while managing diminishing returns. This study proposes a data-driven media mix determination model that integrates supervised machine… Read full abstract & cite →

Media mix determination saturation points audience reach audience exposure

Abstract doi.org/10.14569/IJACSA.2026.0170217 PDF

18

An Ensemble Learning Framework with Metaheuristic Optimization for Credit Card Fraud Detection

Author 1: Agung Nugroho Author 2: Muhtajuddin Danny Author 3: Ismasari Nawangsih

Credit card fraud detection is a major challenge in the financial system due to the characteristics of highly unbalanced data. This study proposes an ensemble learning approach combined with hyperparameter optimization using a Genetic Algorithm to improve the performance of fraud transaction detection. The results of the experiment showed that… Read full abstract & cite →

Fraud detection ensemble learning Genetic Algorithm Random Forest real-time detection

Abstract doi.org/10.14569/IJACSA.2026.0170218 PDF

19

A Lossless Medical Image Compression Framework Using Logic Minimization

Author 1: Swathi Pai M Author 2: Jacob Augustine Author 3: Pamela Vinitha Eric

Medical image compression is an active research area owing to the growth of the volume of medical image data in digital form. A method for lossless compression of medical images was proposed using logic minimization. The grayscale medical image is split into bit planes, and each bit plane is divided… Read full abstract & cite →

Medical image compression lossless compression Quadtree logic minimization bit plane encoding XOR operation gray coding

Abstract doi.org/10.14569/IJACSA.2026.0170219 PDF

20

An Empirical Evaluation of Multivariate Temporal Convolutional Networks with Global Market Indicators for Forecasting the Indonesian LQ45 Index

Author 1: Yohanes Marakub Author 2: Muhammad Zarlis

This study develops a multivariate Temporal Convolutional Network (TCN) framework to forecast the LQ45 stock index using daily time-series data from January 2015 to January 2025. The objective is to examine whether incorporating global market indicators, namely the Volatility Index (VIX), Brent crude oil price, and the Effective Federal Funds… Read full abstract & cite →

Deep learning financial forecasting multivariate forecasting TCN time series analysis

Abstract doi.org/10.14569/IJACSA.2026.0170220 PDF

21

Integrating ABM and GIS for Flood Evacuation Planning: A Systematic Review and Future Direction

Author 1: Kabir Musa Ibrahim Author 2: Abubakar Ahmad Author 3: Noor Akma Abu Bakar Author 4: Mazlina Abdul Majid Author 5: Azamuddin Rahman

This systematic review examines the integration of agent-based modeling (ABM) and Geographic Information Systems (GIS) in flood evacuation planning from 2015 through early 2025. This review aims to systematically evaluate how ABM and GIS have been integrated in flood evacuation research, identify methodological gaps, and propose a structured framework to… Read full abstract & cite →

Agent-based modeling GIS integration flood simulation spatial modeling evacuation dynamics multi-agent systems

Abstract doi.org/10.14569/IJACSA.2026.0170221 PDF

22

A Novel Lightweight Explainable Multilayer Adaptive RNN-Based Intrusion Detection Framework

Author 1: Nidhi Srivastav Author 2: Rajiv Singh

A rapid increase in the instances of cyberattacks has been observed with the expanding digitization. This leads to an urgent and critical need for developing robust intrusion detection systems (IDS) which can identify the occurrences of malicious activities within the network traffic. The present work proposes a novel, explainable, multilayer… Read full abstract & cite →

IDS network security adaptive techniques RNN cybersecurity explainable AI UNR-IDD

Abstract doi.org/10.14569/IJACSA.2026.0170222 PDF

23

Empirical Validation of Learnability Factors in Web-Based AR: Insights from the LEMARK–Hafsa Model Grounded in Kolb’s Experiential Learning Theory

Author 1: Sayera Hafsa Author 2: Mazlina Abdul Majid Author 3: Shafiq Ur Rehman

Augmented Reality in higher education is transforming learning by providing immersive environments that enhance cognitive and motivational engagement. Despite growing interest, there remain limited empirically validated learnability factors that can support future instructional models, such as the LEMARK-Hafsa model. This research attempts to bridge the identified gap through statistically validating… Read full abstract & cite →

Experiential learning theory educational technology predictive validity structural validity AR-based learning factor validation LEMARK–Hafsa model

Abstract doi.org/10.14569/IJACSA.2026.0170223 PDF

24

Automating Computation Independent Model Elicitation in MDA using Task-Oriented Dialogue with In-Context Learning

Author 1: Mohamed EL Ayadi Author 2: Yassine Rhazali Author 3: Mohammed Lahmer

The Computation Independent Model (CIM) is a cornerstone of the Object Management Group's (OMG) Model-Driven Architecture (MDA), capturing business requirements and domain knowledge independent of specific technologies. However, the elicitation of CIM requirements is often a manual, time-consuming, and error-prone process, susceptible to ambiguities inherent in natural language. Traditional Natural… Read full abstract & cite →

MDA CIM Task-Oriented Dialogue (TOD) In-Context Learning (ICL) Large Language Models (LLMs) requirements elicitation Domain-Specific Language (DSL) Artificial Intelligence (AI) Natural Language Understanding (NLU) BPMN

Abstract doi.org/10.14569/IJACSA.2026.0170224 PDF

25

Mapping Research on Artificial Intelligence in Customer Experience: A Bibliometric Analysis

Author 1: Firdaws Hayoun Author 2: Brahim Ouabouch Author 3: Youssef Aatif Author 4: Taoufiq Yahyaoui Author 5: Fatima Zahra El Arbaoui

The purpose of this article is to conduct a comprehensive bibliometric analysis of research on artificial intelligence and customer experience. The study data was extracted from Scopus and Web of Science, focusing on articles published between 2010 and 2025. VOSviewer and Biblioshiny software were used to map the intellectual landscape… Read full abstract & cite →

Artificial intelligence customer experience marketing personalization

Abstract doi.org/10.14569/IJACSA.2026.0170225 PDF

26

AI-Driven Robotic Waste Sorting for Techno-Economic Assessment in Urban Indonesia

Author 1: Ida Nurhaida Author 2: Mohammad Nasucha Author 3: Hari Nugraha

Urban centers in Indonesia are facing increasing pressure in managing municipal solid waste as a result of rapid population growth, rising labor costs, and stricter demands for high-purity recyclable materials. Manual sorting at Material Recovery Facilities has become progressively less efficient and economically burdensome under these conditions. This study presents… Read full abstract & cite →

Smart waste sorting AI vision YOLOv12 computer vision material recovery facility techno-economic assessment urban Indonesia circular economy

Abstract doi.org/10.14569/IJACSA.2026.0170226 PDF

27

A Unified Benchmarking Framework for Offline Handwritten Signature Verification Using Deep Learning Architectures

Author 1: Eissa Alreshidi

Offline handwritten signature verification (OSV) remains a challenging biometric task owing to the subtle variability of genuine signatures and the sophistication of skilled forgeries. This study introduces a unified benchmarking framework for evaluating eight deep learning architectures—CNN Shallow, CNN Deep, ResNet 18, ResNet 34, MobileNetV2, EfficientNet B0, ViT Tiny, and… Read full abstract & cite →

Offline handwritten signature verification Siamese networks contrastive-learning deep learning architectures ResNet MobileNetV2 EfficientNet vision transformers hybrid CNN-transformer models biometric authentication forgery detection embedding separability writer-independent verification

Abstract doi.org/10.14569/IJACSA.2026.0170227 PDF

28

Deep Learning Based Detection of Prostate Cancer in MRI Using Biopsy-Confirmed Ground Truth

Author 1: Samana Jafri Author 2: Gajanan Birajdar

Prostate cancer is one of the most common malignancies in men, and accurate lesion segmentation in magnetic resonance imaging (MRI) is essential for diagnosis, treatment planning, and disease monitoring. Manual delineation by radiologists is time-consuming and subject to interobserver variability. This study presents an automated, deep learning-based framework for 3D… Read full abstract & cite →

Prostate lesion 3D U-Net MRI biopsy confirmed lesion masks

Abstract doi.org/10.14569/IJACSA.2026.0170228 PDF

29

A Composite Approach Extracting Software Quality Attributes Assessing Operational Profiles

Author 1: Sudhakar Kambhampati Author 2: G Vamsi Krishna

In modern software systems, software reliability is a defining quality characteristic, particularly in on-demand and critical needs settings where usage trends are prone to the evolution of time. The traditional black box reliability models do not sufficiently capture these dynamics, especially in component-based and reusable software architectures. In a bid… Read full abstract & cite →

Software reliability operational profile evaluation composite reliability modeling

Abstract doi.org/10.14569/IJACSA.2026.0170229 PDF

30

Robust Medical Image Reconstruction Using a Self-Evolving Encoder–Decoder and Adaptive Convolutional Power Scaling

Author 1: Dhanusha P B Author 2: J. Bennilo Fernandes Author 3: A. Muthukumar Author 4: A. Lakshmi

Robust medical image reconstruction is a critical requirement for accurate diagnosis and clinical decision-making, particularly when images are affected by degradation, noise, or low resolution. Conventional encoder–decoder-based reconstruction methods compress input images into low-dimensional representations and subsequently decode them into high-resolution outputs; however, such approaches often suffer from artifacts and… Read full abstract & cite →

Dynamic encoder and decoder power flex model layer high resolution images weight initialization adaptive convolutional power scaling

Abstract doi.org/10.14569/IJACSA.2026.0170230 PDF

31

Cybersecurity Awareness Among Undergraduate Students in Saudi Arabia: A Quantitative Study

Author 1: Turky A. Saderaldin Author 2: Ali Abuabid

Cybersecurity awareness has become increasingly important as digital services expand rapidly in Saudi Arabia. This study presents a quantitative assessment of cybersecurity awareness among undergraduate students in Riyadh and Jeddah. A structured questionnaire was administered to 177 students to evaluate awareness across three dimensions: internet usage, information security practices, and… Read full abstract & cite →

Cybersecurity awareness undergraduate students smartphone security Saudi Arabia

Abstract doi.org/10.14569/IJACSA.2026.0170231 PDF

32

An Analytical Study of Data Augmentation Across Audio Representations for Infant Cry Classification

Author 1: Meriyem Ghanjaoui Author 2: Abdelaziz Daaif Author 3: Abdelmajid Bousselham Author 4: Sajid Rahim Author 5: Ahmed Bouatmane Author 6: Mohamed Elyoussfi

Several multidisciplinary studies consider an infant’s cry as a valuable source of information, particularly for parents, caregivers, and medical professionals. From a signal processing viewpoint, infant cries can be represented either in the time domain (one-dimensional or 1D raw waveform) or in the time–frequency domain (two-dimensional or 2D spectrogram-based representations)… Read full abstract & cite →

CNN waveform spectrogram deep learning baby cries

Abstract doi.org/10.14569/IJACSA.2026.0170232 PDF

33

Hybrid BERT–BiLSTM Architecture for Enhanced Cyber Threat Intelligence Classification

Author 1: Syarif Hidayatulloh Author 2: Salman Topiq Author 3: Ifani Hariyanti Author 4: Dwi Sandini

Cyber Threat Intelligence (CTI) plays a crucial role in supporting proactive cybersecurity defence by offering insights into adversarial behaviours and attack tactics. However, CTI data are mainly presented in unstructured natural language, characterised by dense technical terminology, implicit attack semantics, and sequential descriptions of multi-stage threat activities. While transformer-based language… Read full abstract & cite →

Cyber Threat Intelligence hybrid deep learning BERT–BiLSTM architecture text classification sequential modelling cybersecurity natural language processing

Abstract doi.org/10.14569/IJACSA.2026.0170233 PDF

34

An Offline Estimation Method for Hip and Knee Joint Angles of Lower Limbs Based on Quaternion and DTW Time Alignment

Author 1: Yuling Zhang Author 2: Yuejian Hua Author 3: Shengli Luo Author 4: Xiaolong Shu Author 5: Hongyan Tang Author 6: Hongliu Yu

Gait analysis is crucial for disease diagnosis and rehabilitation assessment; however, traditional optical motion capture systems are costly and limited to fixed setups. This study presents an "Offline Estimation Method for Hip and Knee Joint Angles of Lower Limbs Based on Quaternion and DTW Time Alignment." The method uses quaternion… Read full abstract & cite →

Inertial measurement unit dynamic time warping lower-limb joint angle gait analysis

Abstract doi.org/10.14569/IJACSA.2026.0170234 PDF

35

Radiomics Feature Profiling of Brodmann Regions in Structural MRI: A Machine Learning Study of Intensive Verbal Memorisation (Huffaz vs Controls)

Author 1: Mohd Zulfaezal Che Azemin Author 2: Iqbal Jamaludin Author 3: Abdul Halim Sapuan Author 4: Mohd Izzuddin Mohd Tamrin

Memorisation-based cognitive training has been hypothesized to relate to experience-dependent brain plasticity; however, quantitative evidence at the regional level remains limited. We hypothesized that radiomics descriptors extracted from Brodmann-area volume-of-interest (VOI) regions in pre-processed structural MRI would contain sufficient information to discriminate Quran memorizers (Huffaz) from non-memorizers (controls), and we… Read full abstract & cite →

Radiomics PyRadiomics Magnetic Resonance Imaging (MRI) voxel-based morphometry (VBM) neuroplasticity Huffaz Brodmann-areas Volume of Interest (VOI) texture analysis machine learning nested cross-validation ROC-AUC Random Forest

Abstract doi.org/10.14569/IJACSA.2026.0170235 PDF

36

Cognitive Assistance for Prosopagnosia Patients Using Landmark-Based Identity and Emotion Recognition

Author 1: Bhavana Nagaraj Author 2: Rajanna Muniswamy

Prosopagnosia is a neurological condition that significantly affects the social interaction and quality of life of individuals. Existing assistive systems mainly focus on either face identity recognition or face emotion recognition, limiting their effectiveness in cognitive assistive scenarios. To overcome these, an integrated framework is needed that jointly addresses identity… Read full abstract & cite →

Cognitive assistance system EfficientNet-B3 emotion recognition face identity recognition prosopagnosia

Abstract doi.org/10.14569/IJACSA.2026.0170236 PDF

37

Comparative Analysis of Neural Network Architectures for Classifying Depressive Content in Social Networks

Author 1: Yntymak Abdrazakh Author 2: Rita Ismailova Author 3: Nurseit Zhunissov Author 4: Arypzhan Aben Author 5: Anuarbek Amanov Author 6: Aigerim Baimakhanova

Depression-related language on social media provides measurable signals for population-level mental-health research, yet model selection remains sensitive to evaluation protocol, domain shift, class imbalance, and computational constraints. This study benchmarks CNN, LSTM, and transformer encoders (BERT, RoBERTa, DistilBERT, and MentalBERT) for binary depression-indicative versus control classification on a unified corpus… Read full abstract & cite →

Depression detection social media text natural language processing cross-platform evaluation robustness statistical significance testing transformer-based models CNN LSTM BERT MentalBERT

Abstract doi.org/10.14569/IJACSA.2026.0170237 PDF

38

Attention-Based Capsule Network with Vision Transformer for Underwater Hyperspectral Image Classification

Author 1: Thiyagarajan B Author 2: Thenmozhi M

The underwater investigations and research remain challenging due to various underwater distortion factors, scattering, and low wavelength absorption. Hyperspectral imaging helps in obtaining detailed information on each underwater object using the spectral reflectance, using 100 to 300 bands. The Hyperspectral imaging in underwater applications contains the 3D hyperspectral cube, which… Read full abstract & cite →

Underwater imaging hyperspectral U-Net Vision Transformer CapsNet classification segmentation

Abstract doi.org/10.14569/IJACSA.2026.0170238 PDF

39

Effectiveness of Capsule Networks in Detecting Deepfakes Instead of Traditional CNNs

Author 1: M. C. Weerawardana Author 2: T. G. I. Fernando

As artificial intelligence has advanced, computer-generated fake content has become increasingly prevalent. Deepfake is an advanced fake creation generated using deep learning-based technologies, and deepfake images, videos, and voices have spread rapidly. The distinction between real and fake content is very hard for the naked eye, making reliable detection essential… Read full abstract & cite →

Deepfake deep learning capsule network convolutional neural network

Abstract doi.org/10.14569/IJACSA.2026.0170239 PDF

40

Machine-Learning–Assisted Probabilistic Wind Assessment at Sechura, Peru

Author 1: Ubaldo Yancachajlla Tito Author 2: Celso Antonio Sanga Quiroz Author 3: Edilberto Velarde Coaquira Author 4: Germán Belizario Quispe

Accurate characterization of wind resources is essential for reliable energy yield estimation and wind farm planning, particularly in regions with limited long-term measurements. This study presents a machine-learning–assisted probabilistic wind assessment in Sechura, Peru, based on multi-year hourly wind data obtained from the NASA POWER database. A representative Typical Meteorological… Read full abstract & cite →

Wind resource assessment probabilistic modeling machine learning kernel density estimation Gaussian mixture model annual energy production capacity factor

Abstract doi.org/10.14569/IJACSA.2026.0170240 PDF

41

Ziraai SSI: A Blockchain-Based Self-Sovereign Identity Model for Agricultural Supply Chains

Author 1: Tariq Alar Author 2: Gnana Bharathy Author 3: Usha Batra Author 4: Mukesh Prasad

The agricultural supply chain plays a critical role in ensuring food security and sustainability; however, it continues to face challenges related to data fragmentation, limited transparency, and insufficient trust among participating stakeholders. Existing supply chain systems are primarily based on centralized identity and data management models, which introduce single points… Read full abstract & cite →

Blockchain self-sovereign identity (ssi) agricultural supply chain verifiable credentials decentralized identity privacy-preserving authentication

Abstract doi.org/10.14569/IJACSA.2026.0170241 PDF

42

Rubric-Relational Discourse Modeling with Counterfactual Explainability for Multi-Trait Automated Essay Scoring

Author 1: N. Sreedevi Author 2: M. Madhusudhan Rao Author 3: Sridevi Dasam Author 4: Roopa Traisa Author 5: Jasgurpreet Singh Chohan Author 6: V. Saranya Author 7: Ahmed I. Taloba

Automated Essay Scoring (AES) systems often rely on holistic prediction and show weak alignment with rubric-based human evaluation. Existing deep learning approaches achieve moderate agreement but struggle to model discourse coherence and provide trait-faithful explanations. This study proposes a rubric-aware and discourse-faithful essay scoring framework that integrates contextual embeddings with… Read full abstract & cite →

Automated Essay Scoring rubric-aware modeling discourse representation counterfactual explainability multi-task learning

Abstract doi.org/10.14569/IJACSA.2026.0170242 PDF

43

Face Sketch Recognition: Ethnic Groups Classification and Recognition Via a VGG16 Model Approach

Author 1: Khalid OUNACHAD Author 2: Mohamed EL GHMARY

In the law enforcement investigation, the police use sketching techniques to identify suspects from an eyewitness's memory. Many automatic face sketch recognition systems that determine the perpetrator’s appearance from the face image datasets have been proposed. The aim is to conduct the arrest of the right offender. We propose this… Read full abstract & cite →

Forensic face sketch ethnic classification deep learning transfer learning optimized VGG16 model

Abstract doi.org/10.14569/IJACSA.2026.0170243 PDF

44

Integrating Japanese Festival Traditions into Computational Thinking: A Culturally Responsive Approach to Micro:Bit-based Physical Computing

Author 1: Daiki Sugiyama Author 2: M. Fahim Ferdous Khan Author 3: Ken Sakamura

The current pre-tertiary educational landscape in Japan is defined by a significant paradox. While technological infrastructure has reached unprecedented levels of saturation through the Global and Innovation Gateway for All (GIGA) School initiative, student engagement and qualitative learning outcomes in computational disciplines have shown signs of stagnation. This research addresses… Read full abstract & cite →

Culturally responsive teaching (CRT) Culturally Responsive Computing (CRC) computational thinking micro:bit K-12 education physical computing STEM education

Abstract doi.org/10.14569/IJACSA.2026.0170244 PDF

45

Machine Learning-Based Effort Prediction and Early Risk Detection in Software Development Projects: A Case Study

Author 1: Andreea-Elena Catana Author 2: Adriana Florescu

Accurate effort estimation and early risk detection are critical for the success of software projects, as inaccurate forecasts can lead to schedule overruns, inefficient resource allocation, and unmet requirements. This study investigates the use of machine learning techniques to support task-level effort prediction and proactive risk identification in software project… Read full abstract & cite →

Software effort estimation risk detection machine learning python open data gradient boosting effort estimation risk thresholding

Abstract doi.org/10.14569/IJACSA.2026.0170245 PDF

46

Self-Supervised and Explainable Transformer-Based Architectures for Robust End-to-End Speech and Language Understanding

Author 1: Mahfuzul Huda

The primary aim of this study is to meld self-supervised learning techniques with transparent transformer-based frameworks to enable resilient, end-to-end speech and language understanding, alongside pretraining deep transformer models using unannotated speech and text corpora. But the system's complicated structure makes it very hard to compute, and its ability to… Read full abstract & cite →

Transformer models self-supervised learning explainable AI speech recognition natural language understanding end-to-end systems

Abstract doi.org/10.14569/IJACSA.2026.0170246 PDF

47

Secure Ultra-Low Latency Data Paths: A Hybrid Architecture for High Speed Networking under Adversarial Conditions

Author 1: Abdulbasid Banga

The need to satisfy the line rate and deterministic latency requirements of next generation industrial networks is imperative for blistering 400G/800G Ethernet backbones, optical transport networks, and breaking new ground in 5G/6G infrastructures. Animals News: Other solutions limit attention towards detection accuracy or throughput separately without providing a discriminative systems-level… Read full abstract & cite →

High Speed Secure Networking Architecture (HSSNA) hardware-accelerated security ai-based anomaly detection 5g/6g network security low-latency encryption pipelines

Abstract doi.org/10.14569/IJACSA.2026.0170247 PDF

48

Hybrid Deep Learning for Academic Achievement Prediction Using Spatio-Temporal and Behavioral Data in Higher Education

Author 1: Asim Seedahmed Ali Osman

Accurate prediction of student academic performance is essential for enabling timely and effective educational interventions. Many existing prediction approaches focus either on academic outcomes or behavioral trends, without fully capturing the interaction between spatial performance indicators and their temporal evolution. To address this limitation, this study proposes a hybrid deep… Read full abstract & cite →

Hybrid deep learning CNN LSTM student performance FOX optimization

Abstract doi.org/10.14569/IJACSA.2026.0170248 PDF

49

Farm and Learn: An Offline Mobile Learning System Integrating AR, AI, and Game-Based Learning for Agricultural Education Among Children

Author 1: Pubuditha De Silva Author 2: Chamodya Prabodhani Author 3: Daminda Herath

This study presents Farm and Learn, an offline-first mobile learning system that integrates Augmented Reality (AR), Artificial Intelligence (AI), and Game-Based Learning (GBL) to enhance agricultural education among children in low-connectivity environments. Existing agricultural learning applications often provide isolated functionalities such as visualization or plant recognition, with limited pedagogical integration… Read full abstract & cite →

Agricultural education augmented reality artificial intelligence child-centered learning game-based learning offline mobile learning

Abstract doi.org/10.14569/IJACSA.2026.0170249 PDF

50

An Enhanced Approach for Workmen’s Compensation Insurance Fraud Detection Based on Fuzzy Rule-Based System

Author 1: Reham M. Essa

In Workmen’s Compensation insurance, fraud detection (FD) remains a significant challenge due to claims' inherent uncertainty and complexity. To address this, we propose an enhanced approach based on a fuzzy rule system (FRS) for FD. The FRS is designed to handle ambiguous and imprecise data, making it effective for identifying… Read full abstract & cite →

Workmen’s compensation fuzzy logic fraud detection rule-based system insurance fraud prediction

Abstract doi.org/10.14569/IJACSA.2026.0170250 PDF

51

Design of a Vision Transformer-Based Architecture for Automatic Facial Emotion Monitoring in Workplace Environments

Author 1: Renzo Sebastian Gonzalez Caceres Author 2: Jeramel Melissa Avila Saldaña Author 3: Patricia Gissela Pereyra Salvador

Facial emotion recognition is increasingly considered in affective computing as a mechanism for unobtrusive emotional awareness in organizational environments. This study proposes the design of a Vision Transformer (ViT)-based system architecture for automatic facial emotion monitoring, focusing on deployability, modular integration, and data-governance considerations rather than model benchmarking. The architecture… Read full abstract & cite →

Facial emotion recognition vision transformer affective computing system architecture design workplace emotion monitoring privacy-by-design deep learning inference

Abstract doi.org/10.14569/IJACSA.2026.0170251 PDF

52

Integrating Acoustic and Image Data Features for Melon Ripeness Classification Using Convolutional Neural Network

Author 1: Endang Purnama Giri Author 2: Agus Buono Author 3: Karlisa Priandana Author 4: Dwi Guntoro

This study evaluates three classification scenarios: image-based only, acoustic-based using Mel Frequency Cepstral Coefficients (MFCC), and a combined multimodal CNN architecture integrating both modalities. The experiments are conducted on a relatively small dataset comprising only 230 samples. To mitigate the risk of overfitting arising from the limited dataset size, data… Read full abstract & cite →

CNN multimodal learning melon ripeness classification image classification acoustic classification data augmentation

Abstract doi.org/10.14569/IJACSA.2026.0170252 PDF

53

Comparative Analysis of Machine Learning Based Algorithms for Predicting Injury Severity in Road Accidents

Author 1: Soumaya AMRI Author 2: Mohammed AL ACHHAB Author 3: Mohamed LAZAAR

Road crash injury severity prediction is essential for intelligent transportation systems, yet challenged by severe class imbalance, rigid 4-class severity schemes (unhurt/slight/hospitalized/fatal), and optimal methodological selection. This study proposes a structured framework systematically evaluating four machine learning models—CatBoost, HistGradientBoosting, Random Forest, and SVM—across multiclass (native 4-class and ordinal wrapper), binary… Read full abstract & cite →

Machine learning imbalanced data road accident multiclass classification binary classification injury severity prediction

Abstract doi.org/10.14569/IJACSA.2026.0170253 PDF

54

Machine Learning and Deep Learning for Detecting Fake News in a Low-Resource Language

Author 1: Elton Tata Author 2: Jaumin Ajdarim Author 3: Nuhi Besimi

Fake news detection has become a major problem in the digital age. This study presents an improved machine learning technique that achieves 91.99% accuracy in predicting fake news detection within Albanian textual datasets, demonstrating an improvement over existing baseline methodology. The implemented learning model uses 54 features that are specific… Read full abstract & cite →

Fake news Albanian language machine learning NLP

Abstract doi.org/10.14569/IJACSA.2026.0170254 PDF

55

AMCS: Adaptive Multi-Controller SDN Security with Stateful Traffic Intelligence for Fast and Accurate Multi-Vector Attack Detection

Author 1: Ameer El-Sayed Author 2: Mohamed Nosseir Hemdan Author 3: Noha Abdelkarim Author 4: Ehab Rushdy Author 5: Hanaa M. Hamza

The rapid proliferation of Software-Defined Networking (SDN) in large-scale Internet of Things (IoT) ecosystems has amplified exposure to sophisticated, multi-vector cyberattacks that simultaneously exploit control- and data-plane asymmetries. Existing single-controller statistical detectors, while effective under high-volume anomalies, fail to sustain precision and responsiveness under dynamic, distributed, or low-rate attack conditions… Read full abstract & cite →

SDN IoT multi-controller security stateful traffic analysis adaptive anomaly detection distributed mitigation entropy-based indicators cooperative defense P4 programmable networks cyber-physical systems security

Abstract doi.org/10.14569/IJACSA.2026.0170255 PDF

56

Wavelet-Based Dual-Domain Phase Alignment for Predicting Stock Index Trends from Investor Sentiment Cycles

Author 1: Monika Gorkhe Author 2: Diksha Tripathi Author 3: Pravin D Sawant Author 4: Elangovan Muniyandy Author 5: Veera Ankalu Vuyyuru Author 6: Pratik Gite Author 7: Raman Kumar

The dynamics in the financial markets are complicated and non-stationary, with a significant influence of the wave of investor sentiment. The recent changes in sentiment-based stock prediction have shown promising results, but the current research is to a large extent based on individual domain analysis, constant correlation, or the traditional… Read full abstract & cite →

Stock trend prediction investor sentiment analysis wavelet time–frequency analysis phase synchronization modeling financial market forecasting

Abstract doi.org/10.14569/IJACSA.2026.0170256 PDF

57

Machine Learning-Based Autism Spectrum Disorder Classification Using an Enhanced Convolutional Neural Network Algorithm

Author 1: P. Yugander Author 2: M. Jagannath

Autism Spectrum Disorder (ASD) is a complex neurological developmental disability that appears during early childhood. Conventional ASD diagnostic techniques rely on behavioural observations, characteristics, and clinical interviews. To overcome these limitations, numerous machine learning (ML) and Deep Learning (DL) techniques have been used to assist physicians. For the past three… Read full abstract & cite →

Autism enhanced CNN random forest MR images logistic regression

Abstract doi.org/10.14569/IJACSA.2026.0170257 PDF

58

Privacy-Preserving Federated Learning for Multi-Institutional Lung Cancer Severity Detection

Author 1: Ch. Srividya Author 2: K. Ramasubramanian Author 3: Myneni.Madhu Bala

Lung cancer continues to be the most common cause of cancer-related mortality globally and the timely detection of lung cancer and classification of its severity levels are critical to improving survival. Nonetheless, data privacy regulations and institutional data silos often create barriers to developing advanced robust AI models among clinical… Read full abstract & cite →

Federated learning privacy-preserving AI lung cancer detection severity classification multi-institutional data

Abstract doi.org/10.14569/IJACSA.2026.0170258 PDF

59

RiskMIS: A Web-Based Risk Management Information System

Author 1: Alya Thiab Almutairi Author 2: Kassem Saleh

Risk management is the systematic process of identifying, assessing, monitoring, and responding to risks in projects and ongoing operations. Effective execution of risk management activities is essential for the successful completion of projects and the achievement of key performance indicators in operational environments. In recent years, organizations have increasingly emphasized… Read full abstract & cite →

Risk management risk management information system risk register project risk decision support systems

Abstract doi.org/10.14569/IJACSA.2026.0170259 PDF

60

Uncertainty-Aware Volumetric Transformer with Dual Spatial-Channel Attention for Lung Nodule Classification

Author 1: B. N. Patil Author 2: TK Rama Krishna Rao Author 3: Nurilla Mahamatov Author 4: Elangovan Muniyandy Author 5: Arun Prasad.VK Author 6: Chamandeep Kaur Author 7: Aaquil Bunglowala Author 8: Ahmed I. Taloba

Lung cancer is also among the most common causes of cancer-related deaths in the world, and the earliest possible detection of the cancer through computed tomography (CT) is important in the enhancement of patient survival. Nevertheless, accurate diagnosis is still a challenge as the nodules are small and indistinct, inter-rater… Read full abstract & cite →

3D Swin Transformer dual-level attention mechanism computed tomography lung cancer diagnosis uncertainty-aware deep learning

Abstract doi.org/10.14569/IJACSA.2026.0170260 PDF

61

An Articulatory-Aware CNN-BiGRU-Attention Framework for Explainable Phoneme-Level Pronunciation Assessment in ESL Speech

Author 1: P. Bindhu Author 2: Jasgurpreet Singh Chohan Author 3: M. Durairaj Author 4: Megha Sawangikar Author 5: N. Neelima Author 6: Elangovan Muniyandy Author 7: G. Sanjiv Ra Author 8: Loay F. Hussien

Proper pronunciation at the phoneme level has been known to be one of the most enduring problems affecting the Second Language learners of the English language (ESL) since the slight pronunciatory variations in the learned language may greatly influence its communicative power and the level of intelligibility. The existing methods… Read full abstract & cite →

Articulatory error analysis attention mechanism ESL pronunciation assessment phoneme recognition model speech processing framework

Abstract doi.org/10.14569/IJACSA.2026.0170261 PDF

62

Cloud-Continuum-Based Deep Learning Optimization Framework for Next-Generation Healthcare Data Performance on IoT Platform

Author 1: G. Aravindh Author 2: K. P. Sridhar

The development of healthcare data performance analysis is becoming more driven by the incorporation of intelligent computing paradigms that guarantee real-time, scalable, and personalized feedback for coaches and athletes. However, existing healthcare data analytics systems are challenged with severe issues such as decision-making latency, processing capacity limitations at the edge… Read full abstract & cite →

Cloud continuum next-generation healthcare data performance platforms fog edge orchestration device

Abstract doi.org/10.14569/IJACSA.2026.0170262 PDF

63

Adaptive Network Security Framework for Distributed Quantum-Assisted Cloud Continuum Architectures

Author 1: P. Suseendhar Author 2: K. P. Sridhar

Distributed Quantum-assisted Cloud Continuum architectures have brought revolutionary changes to the current computing environment through increased data proximity, reduced latency, and real-time responsiveness. These systems are architectures that integrate cloud, fog, and edge computing layers. However, this architectural development comes with many complicated security concerns such as heterogeneous devices, dynamic… Read full abstract & cite →

Adaptive network security distributed cloud continuum threat aware orchestration intelligent engine policy enforcement module

Abstract doi.org/10.14569/IJACSA.2026.0170263 PDF

64

Deep Hybrid Learning for Sustainable Industrial Forecasting: Integrating CNN–LSTM Models to Enhance Economic Efficiency and Carbon Performance

Author 1: Mohamed Amine Frikha Author 2: Mariem Mrad Author 3: Younes Boujelben Author 4: Soufiene Ben Othman

This paper explores the contribution of neural network-based safeguarding models to enhancing the environmental resilience and economic efficiency of industrial supply chains. The methodology includes a review of existing literature for a quasi-experimental study conducted from the perspective of a manufacturer. Using this approach, the study analyzes the transition from… Read full abstract & cite →

Neural networks deep learning CNN-LSTM AI-enabled demand forecasting sustainable supply chain economic performance digital transformation data quality emerging economies

Abstract doi.org/10.14569/IJACSA.2026.0170264 PDF

65

A System-Oriented Machine Learning Approach for Planning and Execution of Decisions in Software Project Management

Author 1: Foziah Gazzawe

Software project management must make high-stakes decisions under uncertainty in effort estimation, cost control, and execution risk. Although machine learning has enhanced predictive accuracy, several studies employ it only in isolated planning or execution tasks, thereby limiting its usefulness as an end-to-end decision support system. This study describes DeciBoost PM… Read full abstract & cite →

Software project management decision support systems machine learning CatBoost effort estimation cost estimation risk prediction issue-tracking explainable AI

Abstract doi.org/10.14569/IJACSA.2026.0170265 PDF

66

Cluster Domain-Aware Client Selection for Federated Learning in The Healthcare Field (CDCSF)

Author 1: Sanaa Lakrouni Author 2: Marouane Sebgui Author 3: Slimane Bah

Client selection remains a critical challenge in Federated Learning (FL). Resource-aware strategies aim to reduce training delays and mitigate stragglers by selecting an appropriate subset of clients in each round. However, these methods prioritize computationally strong clients and exclude resource-constrained clients. In healthcare settings, this approach is impractical because it… Read full abstract & cite →

Federated learning healthcare distributed learning data heterogeneity

Abstract doi.org/10.14569/IJACSA.2026.0170266 PDF

67

Temporal Attention Networks for Real-Time Multimodal Emotion Recognition from EEG and fNIRS Signals

Author 1: Vinod Waiker Author 2: Anne Marie D. Pahiwon Author 3: Ankush Mehta Author 4: Gadug Sudhamsu Author 5: Pavithra M Author 6: K. Kiran Kumar Author 7: B Kiran Bala Author 8: Osama R. Shahin

Emotion recognition is critical in the development of real-time mental health care and individualized cognitive behavior. Current strategies to recognize cognitive emotions frequently fail to capture complex time dependencies and multimodal physiological reactions, leading to sub-optimal performance and inaccurate generalization. To overcome such shortcomings, the proposed study suggests TCADNet, a… Read full abstract & cite →

Emotion recognition Temporal Convolutional Network attention mechanism mental healthcare EEG

Abstract doi.org/10.14569/IJACSA.2026.0170267 PDF

68

Explainable AI with DRL for Smart Home Energy Management and Residential Cost Savings in IoT-Based Autonomic Systems

Author 1: Prabagaran A Author 2: Srinivasan Sriramulu Author 3: S. Premkumar Author 4: Sheelesh Kumar Sharma Author 5: Karthi Govindharaju Author 6: Angelin Blessy J

As more people seek ways to improve their homes and workplaces while reducing energy consumption, smart home systems are becoming increasingly prevalent. Unfortunately, the complexity and "black-box" nature of these systems made it difficult to deploy AI-enabled decision-making simulations, raising issues with explainability, confidence, transparency, responsiveness, and fairness. Explainable Artificial… Read full abstract & cite →

Explainable artificial intelligence smart home energy management deep reinforcement learning Internet of Things energy optimization

Abstract doi.org/10.14569/IJACSA.2026.0170268 PDF

69

Securing Blue Carbon Accounting: A Cryptographic Framework for Coastal Ecosystem Monitoring

Author 1: Heider A. M. Wahsheh

Blue carbon ecosystems have significant long-term carbon sequestration capacity, making them an important nature-based solution for climate change mitigation. However, monitoring and accounting processes are increasingly dependent on distributed IoT sensors, satellite remote sensing, and cloud-based analytics platforms. This growing digitalization exposes the blue carbon data lifecycle to risks such… Read full abstract & cite →

Blue carbon accounting environmental cybersecurity blockchain security Public Key Infrastructure (PKI) elliptic curve cryptography secure multiparty computation IoT security data integrity carbon credit verification MRV

Abstract doi.org/10.14569/IJACSA.2026.0170269 PDF

70

Privacy-Aware Customer Segmentation Using a Distributed Graph-Based Attribute Projection Framework

Author 1: Pentareddy Ashalatha Author 2: G. Krishna Mohan

Customer segmentation plays a vital role in Business Intelligence (BI) by enabling organizations to understand customer behavior, enhance personalization, and support informed decision-making. Conventional segmentation approaches, including K-Means clustering, hierarchical methods, and hybrid deep learning models, often face limitations when handling high-dimensional customer data and typically lack built-in mechanisms to… Read full abstract & cite →

Business intelligence privacy-aware customer segmentation graph-based learning federated analytics graph neural networks

Abstract doi.org/10.14569/IJACSA.2026.0170270 PDF

71

IoT-Driven Sensor Selection for Smart Water and Electricity Systems: A TOPSIS and VIKOR Approach

Author 1: Oumaima Rhallab Author 2: Rachid Dehbi Author 3: Zouhair Ibn Batouta Author 4: Amine Dehbi

In the era of smart cities and intelligent resource management, the need for precise and efficient sensing devices is increasingly critical. Among essential infrastructures are water and electricity systems, which necessitate continuous monitoring to optimize consumption, detect anomalies, and reduce operational costs. Several commercial sensors for water metering and electricity… Read full abstract & cite →

Water and electricity system water metering electricity metering MCDM techniques TOPSIS VIKOR IoT smart metering

Abstract doi.org/10.14569/IJACSA.2026.0170271 PDF

72

A Feasibility Study on Synthetic RGB-NIR Image Generation for Oil Palm Fresh Fruit Bunch Grading

Author 1: Nor Surayahani Suriani Author 2: Norzali Hj Mohd Author 3: Shaharil Mohd Shah Author 4: Siti Zarina Muji Author 5: Fadilla Atyka Nor Rashid

Accurate ripeness of grading oil palm fruit bunches (FFBs) is essential for optimizing oil quality and harvesting decisions. While near-infrared (NIR) imaging provides useful spectral cues for ripeness assessment, its adoption in field conditions is limited by sensor cost and system complexity. This study presents a low-cost alternative by generating… Read full abstract & cite →

Generative AI deep learning U-Net image translation EfficientNet-B0 MobileNetV3

Abstract doi.org/10.14569/IJACSA.2026.0170272 PDF

73

The Acquiring Optimal Models of Random Forest and Support Vector Machine Through Tuning Hyperparameters in Classifying the Imbalanced Data

Author 1: Dwija Wisnu Brata Author 2: Arif Djunaidy Author 3: Daniel Oranova Siahaan Author 4: Samingun Handoyo

Machine learning models most often misclassify the positive class in the dataset with class imbalance. Besides, a sophisticated model involves the hyperparameters that need to be tuned to the optimal values. The study aims to tune hyperparameters of random forest (RF) and support vector machine (SVM) models using 5-fold cross-validation… Read full abstract & cite →

Area under the curve cross-validation folds Matthew's correlation coefficient optimal hyperparameters oversampling technique

Abstract doi.org/10.14569/IJACSA.2026.0170273 PDF

74

K-Nearest Neighbors Algorithm for Short-to-Medium Term Directional Stock Price Forecasting: An Analysis of Thailand’s Banking Sector

Author 1: Passawan Noppakaew Author 2: Parit Wanitchatchawan Author 3: Kanchana Phuhoy Author 4: Natthasorn Seubwong

This research explores the efficacy of a parsimonious K-Nearest Neighbors (KNN) framework for short-to-medium term stock direction forecasting, focusing specifically on the banking sector within Thailand’s SET50 index. Prior preliminary analysis, aimed at determining the optimal prediction horizon, indicated that a 60-day forecast yielded the most effective results, establishing the… Read full abstract & cite →

Stock trend prediction K-Nearest Neighbors technical analysis

Abstract doi.org/10.14569/IJACSA.2026.0170274 PDF

75

Multi-Criteria Methodology for Selecting Communication Protocols in M2M Environments

Author 1: Oleg Iliev

The continuous expansion of Machine-to-Machine (M2M) communication and Internet of Things (IoT) ecosystems has significantly increased the complexity of selecting appropriate communication protocols and data flow management systems. Contemporary M2M deployments operate across heterogeneous functional domains, including sensor networks, transactional systems, and real-time streaming environments, each imposing distinct and often… Read full abstract & cite →

Machine-to-Machine communication Internet of Things multi-criteria decision making AHP TOPSIS communication protocol selection non-functional requirements

Abstract doi.org/10.14569/IJACSA.2026.0170275 PDF

76

Generative AI as a Catalyst for Interoperability and Data-Driven Decision Support in Healthcare Systems of Developing Countries

Author 1: YNSUFU Ali Author 2: MOSKOLAI NGOSSAHA Justin Author 3: AYISSI ETEME Adolphe Author 4: BOWONG TSAKOU Samuel

Interoperability across heterogeneous information systems remains a persistent challenge, particularly in resource constrained contexts where infrastructures are fragmented and data formats remain incompatible. This study introduces a novel methodology that integrates generative Artificial Intelligence (AI) with the urbanization of information systems to enable scalable and seamless interoperability. The approach employs… Read full abstract & cite →

Decision support systems interoperability generative Artificial Intelligence Large Language Model (LLM) heterogeneous data sources information system

Abstract doi.org/10.14569/IJACSA.2026.0170276 PDF

77

Development of Lightweight Residual Convolutional Neural Network for Efficient Facial Emotion Recognition

Author 1: Yelnur Mutaliyev Author 2: Zhuldyz Kalpeyeva

Facial Emotion Recognition (FER) is essential for successful human-computer interaction; however, deploying robust systems on edge devices remains difficult. Recent techniques, such as Vision Transformers (ViTs) and deep ensemble networks, have achieved high accuracy, but suffer from extreme computational overhead and high latency, making them unsuitable for real-time use on… Read full abstract & cite →

Facial Emotion Recognition residual neural networks lightweight convolutional neural networks affective computing facial expression recognition 2013 dataset CPU-optimized architecture pattern recognition

Abstract doi.org/10.14569/IJACSA.2026.0170277 PDF

78

Integrating Deep Reinforcement Learning for Initialization and Adaptive Pheromone Updates in Ant Colony Optimization for UAV Pathing

Author 1: Mohamed A. Damos Author 2: Wenbo Xu Author 3: Abdolraheem Khader Author 4: Ali Ahmed Author 5: Mohammed Al-Mahbashi Author 6: Almuhannad S.Alorfi

Unmanned Aerial Vehicles (UAVs) are indispensable assets for missions in dynamic and complex environments, requiring highly efficient path planning that simultaneously optimizes the often-conflicting objectives of minimizing flight distance, energy consumption, and mission time. While Ant Colony Optimization (ACO) is a recognized and effective metaheuristic for this domain, its performance… Read full abstract & cite →

Deep Reinforcement Learning Ant Colony Optimization adaptive pheromone update UAV pathing

Abstract doi.org/10.14569/IJACSA.2026.0170278 PDF

79

A Context-Aware Hybrid Recommendation Framework for E-Learning Platforms

Author 1: Kaoutar Errakha Author 2: Amina Samih Author 3: Sanaa Dfouf Author 4: Abderrahim Marzouk

E-learning platforms provide learners with ex-tensive digital resources that enable self-paced and location-independent study. However, the overwhelming volume of learning materials offered by a wide range of institutions and content providers makes personalized guidance increasingly essential for effective knowledge acquisition. As a result, recommender systems have become fundamental components of… Read full abstract & cite →

E-learning recommender systems hybrid approach personalized learning course

Abstract doi.org/10.14569/IJACSA.2026.0170279 PDF

80

Efficient CNN-Based Time-Domain Denoising of Impulsive Noise in NB-PLC Systems

Author 1: Wided Belhaj Sghaier Author 2: Fatma Rouissi Author 3: Héla Gassara Author 4: Fethi Tlili

In this study, a convolutional neural network (CNN)-based time-domain denoising approach is proposed to suppress impulsive noise which is considered as the most sever impairments in narrowband powerline communications (NB-PLC). Unlike conventional techniques, such as clipping and blanking, the proposed method does not require prior knowledge of noise statistics. The… Read full abstract & cite →

NB-PLC Middleton Class-A OFDM impulsive noise deep learning CNN

Abstract doi.org/10.14569/IJACSA.2026.0170280 PDF

81

Leveraging Attention Mechanism and Class Weighting for Legal Event Detection in Chinese Text

Author 1: Jinhong Hu Author 2: Shaidah Jusoh

Event detection is an information extraction task that involves extracting specified event types from textual sequences. Currently, most event detection studies focus on English corpora; there is a lack of exploration in other linguistic contexts. Thus, a study on event detection in the Chinese corpus is essential. Sequence-based event detection… Read full abstract & cite →

Event detection information extraction Chinese corpus recurrent neural network

Abstract doi.org/10.14569/IJACSA.2026.0170281 PDF

82

SIFChain: A Decentralized Framework for Secure Storage Sharing and Dynamic Access Control in Virtual Power Plants

Author 1: Xiaochuan Xu Author 2: Xiao Xin Author 3: Jie Liu Author 4: Ruiqi Fang Author 5: Dekai Liu Author 6: Zhixin Li

Virtual Power Plants (VPPs) face significant challenges in secure data management and sharing, including risks of centralized control, single points of failure, and dynamic access requirements among multiple stakeholders. To address these issues, this study proposes SIFChain, a decentralized framework that integrates Hyperledger Fabric, the InterPlanetary File System (IPFS), and… Read full abstract & cite →

Virtual Power Plant access control CP-ABE IPFS blockchain distributed energy privacy protection

Abstract doi.org/10.14569/IJACSA.2026.0170282 PDF

83

AI-Driven Refactoring: Semantic Reconstruction of Domain Models Using LLM Reasoning

Author 1: Mohamed El BOUKHARI Author 2: Nassim KHARMOUM Author 3: Soumia ZITI

This study examines the application of large language models (LLMs) for automating domain layer reconstruction in legacy systems, with a specific focus on a case study involving water consumption management. The process begins with a deliberately disordered JSON representation that conflates domain, application, and infrastructure issues. An LLM, specifically GPT-5.2… Read full abstract & cite →

Domain-Driven Design large language models AI-driven software refactoring legacy systems modernization semantic code analysis architecture reconstruction GPT LLM domain layer reconstruction AI-assisted software engineering

Abstract doi.org/10.14569/IJACSA.2026.0170283 PDF

84

Enhancing Telecom Churn Prediction Using Emotion-Driven and Behavioral Engagement Features

Author 1: Huthaifa Aljawazneh

Accurate churn prediction enables service providers to develop effective retention strategies and promotes revenue stability in the telecommunication industry. This study enhances churn prediction performance by extracting five emotion-driven and behavioral engagement features from a telecom churn dataset. The new features represent derived, experience-oriented indicators constructed from operational usage data… Read full abstract & cite →

Customer churn prediction emotion-driven features feature engineering imbalanced data classification

Abstract doi.org/10.14569/IJACSA.2026.0170284 PDF

85

Coronary Heart Disease Prediction Using Machine Learning Algorithms

Author 1: Inooc Rubio Paucar Author 2: Cesar Yactayo-Arias Author 3: Laberiano Andrade-Arenas

Cardiopathy is one of the most serious diseases worldwide with its high morbidity and mortality rates posing a latent risk over time. The objective of this research focuses on evaluating Machine Learning (ML) models such as Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Logistic Regression (LR) for the prediction… Read full abstract & cite →

Cardiovascular disease machine learning prediction random forest XGBoost

Abstract doi.org/10.14569/IJACSA.2026.0170285 PDF

86

Integrating Processing-In-Memory into HW/SW Co-Design for Automotive Embedded Systems

Author 1: Zineb El Kacimi Author 2: Safae Dahmani Author 3: Oussama Elissati Author 4: Mouhcine Chami

Continuous expansion of using intelligent learning workloads in modern vehicles, leads Electronic Control Units (ECUs) to manage massive volumes of data while dealing with hard real-time constraints. That said, ECUs must operate under strict power budget due to limited battery capacity and other safety and functional requirements. We study in… Read full abstract & cite →

Electronic Control Units energy consumption hardware/software co-design Processing-In-Memory real-time constraints

Abstract doi.org/10.14569/IJACSA.2026.0170286 PDF

87

A Feasibility Study of Explainable Machine Learning on Small-Scale Postoperative Voice Data

Author 1: Noura Haddou Author 2: Najlae Idrissi Author 3: Sofia Ben Jebara

Voice dysfunction is a common complication following thyroid surgery. However, the application of explainable machine learning for predicting postoperative voice recovery remains largely unexplored. Therefore, an investigation was done to examine voice recovery based on acoustic, objective, and glottal features. Voice recordings were collected from female patients before surgery and… Read full abstract & cite →

XAI explainable AI SHAP glottal features SVM thyroidectomy voice recovery

Abstract doi.org/10.14569/IJACSA.2026.0170287 PDF

88

A Dual-Chain and Differential Privacy-Based Solution for Medical Data Privacy Protection and Access Control

Author 1: Cen Gu Author 2: Luping Wang Author 3: Hongjie Wu

As living standards rise, people are paying increasing attention to health. Vast quantities of medical data are generated daily, yet each piece contains sensitive information such as patients’ names, mobile numbers, email addresses, and places of employment. Should this information be compromised, the consequences would be irreversible, causing severe damage… Read full abstract & cite →

Blockchain differential privacy IPFS medical data

Abstract doi.org/10.14569/IJACSA.2026.0170288 PDF

89

Leveraging Statistical Invariants to Fortify CNN-Based Medical Diagnostics Against Adversarial Perturbations

Author 1: Yassine Chahid Author 2: Anas Chahid Author 3: Ismail Chahid Author 4: Aissa Kerkour Elmiad

The integration of artificial intelligence (AI) in medical diagnostics is increasingly jeopardized by adversarial attacks—imperceptible perturbations designed to induce misclassification in Deep Learning models. While Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance in medical image analysis, their susceptibility to gradient-based attacks poses a severe risk to patient safety and… Read full abstract & cite →

Adversarial defense medical X-ray synergistic ensemble deep learning security diagnostic robustness PGD attack

Abstract doi.org/10.14569/IJACSA.2026.0170289 PDF

90

A Transformer-Based Approach for Multimodal Arabic Sentiment Analysis

Author 1: Ayoub BEN CHEIKHI Author 2: EL Habib NFAOUI

The Multimodal Sentiment Analysis (MSA) land-scape for Arabic content is strikingly underexplored, mainly due to limited datasets and a lack of robust integration methods across text, audio, and image. While transformer-based models like MarBERT and ArBERT achieve strong results on Arabic text, most research remains unimodal and does not fully… Read full abstract & cite →

Arabic sentiment analysis multimodal learning feature fusion machine learning early fusion

Abstract doi.org/10.14569/IJACSA.2026.0170290 PDF

91

HQ-RTVF: High-Quality Real-Time Virtual Try-On Fitting for Diverse Clothing and Body Morphologies

Author 1: Ilham KACHBAL Author 2: Khadija Arhid Author 3: Said El Abdellaoui

The ability to virtually try on clothing items has become an increasingly important feature for e-commerce and online shopping experiences. Real-time virtual try-on remains challenging because existing methods force a trade-off between speed and quality GAN-based approaches achieve high visual fidelity but at low frame rates, while faster methods sacrifice… Read full abstract & cite →

Virtual try-on diffusion models real-time processing deep learning garment synthesis pose estimation

Abstract doi.org/10.14569/IJACSA.2026.0170291 PDF

92

Density-Guided Adaptive Patch Learning for Robust Crowd Counting

Author 1: Abdullah N Alhawsawi

Accurate crowd counting in real-world scenes re-mains challenging due to severe occlusions, perspective distortion, and large intra-scene density variation. Recent deep learning based approaches typically address these challenges using patch-level learning, where images are divided into fixed grids or randomly cropped patches. These approaches then estimate the count in each… Read full abstract & cite →

Computer vision deep learning crowd counting

Abstract doi.org/10.14569/IJACSA.2026.0170292 PDF

93

Forward Selection for Time Series-Based Qubit Generation via Parameterized Quantum Gates

Author 1: Singaraju Srinivasulu Author 2: Nagarajan G

Quantum data processing requires classical data to be encoded into quantum states. Current noisy intermediate-scale quantum devices have a limited number of qubits that are stable only briefly. Encoding classical data into qubits is the initial step in Quantum Machine Learning (QML), and effective encoding is crucial for quantum processing… Read full abstract & cite →

Quantum bits Quantum Machine Learning quantum algorithms quantum measurements Parameterized Quantum Gates feature extraction time-series data

Abstract doi.org/10.14569/IJACSA.2026.0170293 PDF

94

A Hybrid DMD–TCN Framework for Interpretable Short-Horizon Prediction of 6-DOF Ship Motions

Author 1: Enock Tafadzwa Chekure Author 2: Kumeshan Reddy Author 3: John Fernandes

Accurate short-horizon prediction of six degrees of freedom (6-DOF) vessel motions is essential for autonomous navigation, motion compensation, and operational decision making. Traditional seakeeping models rely on hydrodynamic coefficients that are seldom available for full-scale vessels, while purely data-driven approaches may struggle to maintain physical consistency. This study introduces a… Read full abstract & cite →

Dynamic Mode Decomposition Temporal Convolutional Network hybrid learning seakeeping and vessel response data-driven modelling Koopman operator methods

Abstract doi.org/10.14569/IJACSA.2026.0170294 PDF

95

SecureDML:An Intelligent Framework for Preventing Poisoning Attacks in Distributed Machine Learning Systems

Author 1: Archa A. T Author 2: Kartheeban K

The security and protection of models in dis-tributed machine learning (ML) systems require high emphasis on adversarial threats, including poisoning attacks. This study contains a complete framework that integrates different advanced techniques to monitor poison attacks and prevent such attacks for the effective functioning of machine learning systems. The proposed… Read full abstract & cite →

Poisoning attacks SHapley Additive exPlanations anomaly detection federated learning autoencoders

Abstract doi.org/10.14569/IJACSA.2026.0170295 PDF

96

Post-Quantum Module Learning with Rounding-Based Public Key Encryption Using Incomplete Number Theoretic Transform

Author 1: Anupama Arjun Pandit Author 2: Arun Mishra

Post Quantum Cryptographic (PQC) techniques are widely used in encryption standards and digital signatures. Lattice-based post-quantum cryptographic techniques have been reported in the last decades. The present work proposes an optimized quantum-safe lattice public key encryption (PKE) scheme based on the Module Learning with Rounding (MLWR) problem, enhanced by the… Read full abstract & cite →

Post-quantum cryptography public key encryption lattice-based cryptography Incomplete Number Theoretic Transform Module Learning with Rounding Learning with Errors

Abstract doi.org/10.14569/IJACSA.2026.0170296 PDF

97

Abnormal State Detection of Industrial Tools Based on the MGC-YOLOv8 Algorithm

Author 1: Guan Yang Author 2: Xiang Cheng Author 3: Miao Wang Author 4: Ziyue Huang Author 5: Hao Tang Author 6: Yujun Chen

As intelligent manufacturing advances toward precision and automation, cutting tool condition critically impacts product quality, equipment safety, and production efficiency. Anomalies like wear, chipping, or fracture cause workpiece scrapping and machine failure, demanding efficient online monitoring. Traditional manual or image-based methods suffer from low accuracy in complex environments. Although deep… Read full abstract & cite →

Object detection surface defect detection YOLOv8

Abstract doi.org/10.14569/IJACSA.2026.0170297 PDF

98

Tomato Maturity Analysis: A Comparative Study of Detection and Instance Segmentation Using YOLOv8

Author 1: Salma Ait Oussous Author 2: Rachid El Bouayadi Author 3: Driss Zejli Author 4: Aouatif Amine

The accurate visual analysis of fruit maturity in complex agricultural scenes remains a fundamental challenge due to gradual appearance changes, object overlap, and partial occlusion. This study addresses tomato maturity analysis, formally defined as instance-level binary classification and spatial localization under varying degrees of visual density. While bounding-box-based object detection… Read full abstract & cite →

Tomato maturity detection computer vision object detection instance segmentation image analysis Deep Learning

Abstract doi.org/10.14569/IJACSA.2026.0170298 PDF

99

Time Series Anomaly Detection Based on Entropy-Sparsified Time-Frequency Fusion and MsRwGWO Meta-Optimization

Author 1: Xiaogang Yuan Author 2: Jiaxi Chen Author 3: Dezhi An Author 4: Jianxin Wan

Addressing the core challenges in multivariate time series anomaly detection within complex industrial environments, such as redundant time-frequency feature fusion, significant noise interference, and difficulties in model hyperparameter tuning, this study proposes a detection framework (TFUL) based on entropy-sparsified time-frequency fusion and a Multi-strategy Random Weighted Grey Wolf Optimizer (MsRwGWO)… Read full abstract & cite →

Time series anomaly detection entropy sparsification time-frequency fusion Mixture of Experts (MoE) meta-heuristic optimization

Abstract doi.org/10.14569/IJACSA.2026.0170299 PDF

100

Blockchain-Based Multi-Chain Data Supervision Mechanism for Traditional Chinese Medicine Traceability System

Author 1: Rongjun Chen Author 2: Yun Sun Author 3: Feng Xue Author 4: Yongzhi Ma Author 5: Xinyu Wu Author 6: Xianxian Zeng Author 7: Jiawen Li Author 8: Jinchang Ren

Addressing the challenges of Traditional Chinese Medicine (TCM) traceability systems, including heavy data storage burdens, poor privacy protection, and susceptibility to tampering, this study establishes a highly secure and trustworthy traceability supervision system for the entire Chinese medicine supply chain, which enhances product quality and safety assurance. Centred on the… Read full abstract & cite →

Blockchain traceability multi-chain architecture Hyperledger Fabric Traditional Chinese Medicine

Abstract doi.org/10.14569/IJACSA.2026.01702100 PDF

101

Attention-Guided Fusion of EfficientNet-B0 and Swin Transformer for Cervical Cancer Classification

Author 1: Twisibile Mwalughali Author 2: Emmanuel C. OGU Author 3: Evason Karanja

The interpretation of colposcopy images is a critical yet subjective component of cervical cancer screening. To enhance this process, we propose a novel hybrid deep learning framework for the classification of cervical lesions. Our model integrates EfficientNet-B0, adept at extracting localized hierarchical features, with a Swin-Tiny Transformer, which excels at… Read full abstract & cite →

Cervical cancer classification deep learning model colposcopy cross-attention fusion EfficientNet Swin Transformer

Abstract doi.org/10.14569/IJACSA.2026.01702101 PDF

102

Optimizing Resource Allocation for Crisis-Resilient Healthcare Robotics: An Integrated MLR, MDP, and Petri Net Approach

Author 1: Ikram Dahamou Author 2: Ayoub Elbazzazi Author 3: Cherki Daoui

Global crises, such as pandemics and climate-related disasters, place unprecedented strain on healthcare Systems, exposing weaknesses in resource management and patient care. This study aims to address these challenges by developing an integrated computational framework for crisis-resilient healthcare robotics. We propose a unified approach that combines Multiple Linear Regression (MLR)… Read full abstract & cite →

Healthcare robotics Markov Decision Processes Petri Nets Multiple Linear Regression crisis management ICU resource allocation

Abstract doi.org/10.14569/IJACSA.2026.01702102 PDF

103

Enhanced Ant Colony Optimization for Capacitated Vehicle Routing Problem with Time Windows in Franchise Distribution

Author 1: Dian Rachmawati Author 2: Tommy Lohil Author 3: Jos Timanta Tarigan

Efficient routing for distributing goods to multiple franchisee locations requires optimization techniques capable of handling vehicle capacity limits, heterogeneous time windows, and operational constraints, making conventional brute-force or map-based approaches infeasible due to the NP-hard nature of the problem. This study presents an enhanced Ant Colony Optimization (ACO) algorithm for… Read full abstract & cite →

Ant Colony Optimization CVRPTW heuristics distribution routing logistics optimization

Abstract doi.org/10.14569/IJACSA.2026.01702103 PDF

104

Designing an Attack-Vector-Based Taxonomy for IoT Malware

Author 1: Huda Aldawghan Author 2: Mounir Frikha

This study presents a literature-derived, attack-vector-based taxonomy for IoT malware and complements it with an empirical validation using supervised machine learning. Building on prior surveys and taxonomies of IoT security and malware behavior, we synthesize how existing studies implicitly or explicitly describe infection vectors such as credential abuse, exposed services… Read full abstract & cite →

IoT security malware taxonomy attack vectors cyber threat intelligence network defense

Abstract doi.org/10.14569/IJACSA.2026.01702104 PDF

105

Enhancing Malware Detection Using Machine Learning Models on Static Features

Author 1: Ashwag Alotaibi Author 2: Mounir Frikha

This research introduces a CPU-optimized static malware-detection framework for resource-constrained environments, such as endpoints and IoT devices. We address the significant challenge of high memory and computational demands by proposing a robust, memory-safe data ingestion pipeline. This pipeline exclusively extracts histogram-based static features, employs type compression, and utilizes batch-wise loading… Read full abstract & cite →

Malware detection machine learning (ML) static features stacking ensemble CPU optimization resource constraints memory efficiency computational efficiency

Abstract doi.org/10.14569/IJACSA.2026.01702105 PDF

106

Stability-Aware QUBO Feature Selection for Tabular Classification Under Repeated Nested Cross-Validation

Author 1: Marco Fidel Mayta Quispe Author 2: Leonid Alemán Gonzales Author 3: Charles Ignacio Mendoza Mollocondo Author 4: Nayer Tumi Figueroa Author 5: Juan Carlos Juarez Vargas Author 6: Godofredo Quispe Mamani

Quadratic Unconstrained Binary Optimization (QUBO) provides a principled framework for feature selection by encoding relevance–redundancy trade-offs and explicit constraints directly in a combinatorial objective. This study presents a stability-aware QUBO pipeline for tabular binary classification, evaluated on two standard benchmarks, namely Breast Cancer Wisconsin Diagnostic (569 samples, 30 features) and… Read full abstract & cite →

Feature selection QUBO simulated annealing nested cross-validation selection stability Jaccard similarity probability calibration tabular classification

Abstract doi.org/10.14569/IJACSA.2026.01702106 PDF

107

Attention-Guided Bidirectional Temporal Modelling with Graph-Based Regional Spatial Context for Bajra Crop Yield Prediction

Author 1: Mamta Kumari Author 2: Suman Author 3: Devendra Prasad

Bajra (pearl millet) is a very important crop in Rajasthan, India, since it is drought-resistant, nutritious, and culturally important. But its productivity is becoming vulnerable to changes in climate, such as erratic rain and temperature changes, and thus precise estimation of yield is vital. Crop Yield Prediction (CYP) indicators like… Read full abstract & cite →

Bajra crop yield prediction regional crop yield fore-casting attention-guided Bidirectional LSTM saline and alkaline soil composition

Abstract doi.org/10.14569/IJACSA.2026.01702107 PDF

108

Scalability of Predictive Models on Multi-Core CPUs and GPUs: An Empirical Analysis

Author 1: Atif Mahmood Author 2: Wan Joe Dean Author 3: P. Ganesh Kumar Author 4: Adnan N. Qureshi

The scalability of predictive models has become a critical factor in modern machine learning, as data volumes grow and computational resources diversify. This study presents an empirical benchmark of three widely used regression paradigms: Elastic Net, XGBoost, and Multi-Layer Perceptrons (MLPs). The Obesity Estimation dataset is used to evaluate both… Read full abstract & cite →

Scalability XGBoost Neural Networks Elastic Net resource efficiency

Abstract doi.org/10.14569/IJACSA.2026.01702108 PDF