Asian Journal of Research in Computer Science

Task Scheduling in the Fog to Cloud Continuum for IoT Services: A Taxonomy and Structured Synthesis of Distributed Resource Management

Albahlool Abood — 2026-07-21

Task scheduling has often been treated as a secondary concern in fog computing, something to address only after the architecture is defined. This review argues that it is instead the central runtime decision in the fog to cloud continuum, because it determines whether the promised gains in latency, energy efficiency, and reliability can actually be achieved. The study synthesizes 102 foundational, methodological, and technical sources on task scheduling in fog enabled IoT environments. The aim was not simply to catalogue algorithms, but to examine how the field has framed the scheduling problem and how that framing has changed over time. The evidence reveals a clear progression. Early studies commonly assumed stable resources, predictable workloads, and simplified network conditions, which made scheduling easier to model but less representative of real deployments. More recent work has relaxed these assumptions and introduced dynamic, multi objective, application aware, learning based, and deployment oriented approaches. Six research streams emerge from this evolution. The main finding is that algorithmic sophistication has advanced faster than evaluation practice. Reported improvements in latency, energy consumption, and other QoS metrics are often difficult to compare because studies use different workloads, simulators, baselines, and experimental assumptions. Scheduling and orchestration overhead is rarely measured, while physical testbed validation remains limited. These gaps directly affect confidence in whether a proposed scheduler would behave as expected in operational fog systems. The review therefore identifies several priorities for future work: standardized benchmark workloads, cloud native scheduling that accounts for container lifecycle and microservice dependencies, resilience aware scheduling that treats failures and migration as first class concerns, and carbon aware orchestration that extends beyond energy minimization. Beyond the taxonomy, the paper argues for a shift from proof of concept scheduling studies toward reproducible, transparent, and deployable fog systems.

Governance Frameworks for Ensuring Algorithmic Decision Data Integrity in AI Systems and Regulatory Compliance

Pelumi Damola Adeyinka — 2026-07-21

Artificial intelligence systems increasingly mediate consequential decisions in credit allocation, healthcare triage, employment screening and public administration, yet the data underpinning these decisions is frequently incomplete, mislabelled, stale or quietly altered as it moves through long and opaque pipelines. This review examines governance frameworks intended to preserve the integrity of algorithmic decision data and to align organisational practice with an increasingly dense regulatory landscape spanning the European Union, the United States and international standard-setting bodies. It synthesises literature on data quality theory, documentation artefacts such as datasheets and model cards, blockchain-based provenance mechanisms, algorithmic auditing regimes and sector-specific compliance obligations in finance and healthcare. The review finds that technical solutions for data quality monitoring have matured considerably faster than the institutional arrangements needed to make such monitoring auditable, contestable and legally enforceable, producing a persistent gap between what is technically feasible and what is organisationally practised. It further finds that regulatory instruments, notably the General Data Protection Regulation and the Artificial Intelligence Act, converge on transparency and documentation obligations but diverge on enforcement mechanics, creating compliance friction for organisations operating across jurisdictions. The review proposes a layered governance model integrating data-level controls, documentation practices, human oversight and independent auditing, and identifies future research priorities around interoperable provenance standards, cross-border regulatory harmonisation and the measurement of data integrity as a continuous rather than a point-in-time property.

Institutional Apathy and Security Fatigue in Information Privacy and Data Security Governance: A Critical Narrative Review of Psychological Limits and Organisational Design

Onyii Henry — 2026-07-29

Information privacy and data security governance increasingly depend on sustained human attention: employees must interpret warnings, follow changing policies, report anomalies, manage credentials, make disclosure decisions and repeatedly consent to data practices. Yet governance arrangements commonly treat attention, motivation and self-control as effectively unlimited. This critical narrative review examines how security fatigue, privacy fatigue, habituation, burnout, cynicism, organisational silence and institutional decoupling interact to weaken protective behaviour and governance legitimacy. Literature published from 1 January 2000 to 20 May 2026 was identified through accessible scholarly indexes, bibliographic databases, disciplinary digital libraries, DOI metadata services, institutional repositories and citation chaining, with foundational earlier works retained where conceptually necessary. Evidence was appraised for design quality, behavioural measurement, temporal ordering, ecological validity, theoretical coherence and relevance to organisational governance. The synthesis indicates that fatigue is not adequately explained as individual carelessness. Repeated low-value warnings, work-impeding controls, opaque privacy choices, excessive policy demands and punitive reporting climates create cumulative cognitive and emotional costs. These costs can produce attentional habituation, rational workarounds, reduced self-efficacy, resignation and silence. At institutional level, audit-oriented programmes may become decoupled from operational risk reduction, allowing training completion, policy acknowledgement and nominal consent to substitute for observed protective outcomes. Evidence is strongest for warning habituation, compliance-cost reasoning, privacy concern–behaviour discrepancies and associations between exhaustion and silence. Confidence is lower regarding long-term causal pathways and the effectiveness of organisation-wide interventions because much of the literature is cross-sectional, self-reported and based on behavioural intention. An integrated burden–efficacy–legitimacy cycle is proposed to explain how security demands, perceived control, organisational credibility and voice climate jointly shape behaviour. Sustainable governance should reduce unnecessary security work, prioritise high-consequence actions, automate where safe, design adaptive warnings, preserve meaningful choice, support non-punitive reporting and evaluate real behaviour and risk outcomes rather than ceremonial indicators.

Development of Rechargeable Variable Laboratory Bench with Direct Current Power Supply System

Samson Dauda Yusuf — 2026-07-20

Unreliable grid electricity constrains laboratory teaching and research activities in many Nigerian institutions because conventional direct current power supplies depend on continuous mains availability. This study designed and developed a rechargeable variable laboratory bench with an integrated DC power supply and wireless monitoring system. The design provides an adjustable 0–24 V DC output, a maximum current range of 0–5 A, and a seven-cell lithium-ion battery pack rated at 25.9 V nominal and 29.4 V when fully charged. The system comprises an 18 V buck–boost output, 3.3 V and 5 V regulated outputs with a USB port, and a variable 2–24 V output. An Arduino Nano acquires voltage measurements through divider networks, while an ESP8266 module transmits the readings wirelessly for real-time monitoring. The design process involved circuit analysis, Proteus 8.0 simulation, hardware construction, and performance evaluation. Simulation results showed outputs of 12.00 V at the variable stage, 5.01 V at the 5 V rails, 3.30 V at the Wi-Fi rail, and 18.10 V at the dedicated 18 V channel. The charging section produced approximately 30.3 V DC from the stepped-down and rectified mains input. The monitoring algorithm simultaneously reported the three output channels. Under the reported test conditions, the system demonstrated stable voltage regulation, battery-backed operation, protective functionality, and reliable wireless monitoring. Further load, thermal, ripple, current-capacity, and long-duration battery tests are required to establish its full laboratory rating.

Explainable Machine Learning for Career Path Prediction of Student Scholars Using Educational Data Mining

Maricris M. Usita — 2026-07-20

Career motivation is influenced by a variety of academic experiences, family influences, socioeconomic factors, and individual characteristics. Understanding how these factors drive students' career decisions is important for developing effective career guidance programmes and enhancing graduate employability. This study uses an explainable machine learning method to predict the desired career paths of scholarship students at Occidental Mindoro State University, Philippines, based on Educational Data Mining (EDM) techniques. A total of 1,300 student scholars were surveyed, and the data included demographic, academic, socioeconomic, family, and career-development variables. Five supervised learning algorithms—Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and Artificial Neural Network (ANN)—were developed and compared using accuracy, precision, recall, F1-score, and a confusion matrix. Random Forest feature importance and SHapley Additive exPlanations (SHAP) were used to identify influential features and analyze feature interactions, thereby improving model interpretability. Random Forest was the best-performing model, with an accuracy of 97.31%, precision of 97.40%, recall of 97.31%, and F1-score of 97.29% among the evaluated models. The explainability analysis revealed that the factors most strongly associated with students’ career choices were academic program and parents’ career expectations; extracurricular activities, mathematics achievement, family income, and academic performance were also associated with the prediction results. The SHAP interaction results indicated meaningful interactions among academic variables, emphasizing the interrelated nature of career decision-making. The findings underscore the value of integrating predictive analytics and XAI to support transparent, data-informed career counselling and educational planning for higher education students. The proposed framework provides a practical tool to inform career preferences and student development programmes.

An AI-Powered Hybrid Framework for Career Readiness: Job Role Prediction, Skill Gap Analysis, And Personalized Learning Path Recommendation

Vihansa Thathsiluni Chandrakumara — 2026-07-22

Rapid digital transformation and evolving skill requirements have intensified career uncertainty and skill mismatches among information technology job seekers. Existing career-guidance tools often provide fragmented and static support, limiting personalised and future-oriented career planning. This study presents an AI-powered career-readiness platform designed to support job seekers in the information technology domain through integrated job-role prediction, skill-gap analysis, future skill-demand forecasting, learning-path recommendation, and resume optimisation. The system was developed as a web-based platform using a hybrid machine-learning and natural language processing architecture. A hybrid classifier based on a support vector machine and random forest was trained using an IT-domain resume dataset containing 10,174 records and 38 predefined job roles. The platform also incorporated text processing, semantic-similarity analysis, skill forecasting, and large language model-based resume feedback to generate personalised career-guidance outputs. Five publicly available datasets supported resume classification, job-role mapping, skill extraction, course recommendation, and skill-demand forecasting. The classifier was evaluated using an 80:20 stratified train-test split and five-fold stratified cross-validation. It achieved 99.90% accuracy and a weighted F1-score of 99.91% on the test set. Cross-validation produced a mean validation F1-score of 99.85%, with low variation across folds. Independent validation using a limited sample achieved 85% top-1 and 100% top-5 job-role accuracy, supporting real-world generalisation. Through a unified interface, the platform provides predicted job roles, identified missing skills, future skill trends, relevant learning pathways, and resume-improvement suggestions. The findings indicate that an integrated AI-based framework can provide structured and personalised career-readiness support for IT job seekers, although broader validation using diverse real-world datasets remains necessary.

Convolutional Neural Networks and VGG16 Models in Object Detection

Anas Tukur Balarabe — 2026-07-22

Despite the growing need for object-detection systems, companies, governments, industries, and organisations face significant challenges in implementing them to address real-world problems. Object detection supports applications ranging from quality control and inventory management to robotics, surveillance, autonomous systems, and defect detection. Advances in deep learning have enabled notable progress through architectures such as convolutional neural networks (CNNs). This study examined the performance of two deep-learning architectures for detecting household items under controlled experimental conditions. A dataset of 2,100 images, comprising three balanced classes—bottles, boxes, and cups - was annotated and used in the experiment. The results showed that the fine-tuned VGG16 model consistently outperformed the plain CNN across the principal performance measures. Intersection over Union (IoU) increased from 57.11% for the plain CNN to 98.00% for the fine-tuned model, while loss decreased from 6.40% to 4.21%. The fine-tuned model also produced a smoother and more stable learning curve, indicating more consistent detection performance. However, this improvement required a longer training time: approximately 14,400 seconds for the fine-tuned model compared with 12,240 seconds for the plain CNN. The comparison was conducted using the same dataset, training duration, and evaluation framework for both architectures. These findings indicate a trade-off between computational efficiency and predictive performance under the experimental conditions used.

A Weighted Voting Ensemble Machine Learning Framework for Customer Lifetime Value Prediction in E-commerce

Seun Ebiesuwa — 2026-07-23

Customer lifetime value (CLTV) prediction supports the allocation of marketing resources and the identification of high-value customers in e-commerce. Traditional approaches may not adequately represent non-linear and multidimensional purchasing behaviour, particularly when customer-value classes are imbalanced. This study developed a weighted soft-voting ensemble for CLTV tier classification using the Dunnhumby Complete Journey dataset. Transactional, product, and household-demographic data were integrated, and customer-level recency, frequency, monetary, behavioural, promotional, temporal, and demographic features were engineered from the training period. A chronological split was applied at day 533 so that the training data preceded the evaluation period, thereby reducing the risk of temporal leakage. The Synthetic Minority Over-sampling Technique was applied only to the training data. The ensemble combined Random Forest, XGBoost, LightGBM, and CatBoost after isotonic probability calibration. Dynamic voting weights were derived from F1-score, receiver operating characteristic area under the curve, and accuracy. The proposed ensemble achieved 77.80% accuracy, 77.25% macro F1-score, 77.70% precision, 77.33% recall, and 90.58% ROC-AUC. It exceeded the individual models in accuracy, macro F1-score, precision, and recall, although CatBoost produced a slightly higher ROC-AUC. The findings indicate that calibrated, performance-weighted probability aggregation can provide a modest improvement in balanced CLTV tier classification. Further validation across additional retail datasets and market settings is required before broader generalisation.

Deep Learning-based Software Vulnerability Severity Prediction Using Natural Language Processing

Paul Teye — 2026-07-24

Software vulnerability assessment is essential for prioritising remediation, yet the manual assignment of Common Vulnerability Scoring System metrics is time-consuming, subjective, and prone to error. This study develops and evaluates a design science artefact for predicting vulnerability severity directly from natural-language descriptions. More than 15,000 confirmed vulnerability records were collected from ten open-source projects represented in the CVE-NVD repositories. The descriptions were cleaned through punctuation removal, tokenisation, stopword removal, and Porter stemming. Textual features were represented using n-grams, term frequency-inverse document frequency, and Word2Vec, while Information Gain was applied for feature selection. A deep neural network with three hidden layers was developed for binary classification of vulnerabilities as Severe or Non-Severe. Its performance was compared with Random Forest, Support Vector Machine, Decision Tree, k-nearest neighbours, and Naïve Bayes classifiers using stratified 10-fold cross-validation. The deep neural network achieved average accuracy, precision, and recall values of 71.2%, 68.5%, and 72.0%, respectively, and generally outperformed the conventional classifiers across the reported metrics. However, performance varied among projects, with AUC-ROC values ranging from 0.385 for Gentoo to 0.720 for Windows 7, and the Decision Tree exceeded the deep neural network on the Gentoo dataset. These findings indicate that natural-language descriptions can support automated severity prediction, while also showing that model performance depends on project-specific data characteristics and data availability.

An Intelligent IoT-Enabled Automated Drip Irrigation System with Real-time Water Level Monitoring and Adaptive Control

K R Shwetha — 2026-07-25

Traditional irrigation systems can involve substantial water wastage, uneven water distribution, and high labour requirements. This study developed an intelligent Internet of Things (IoT)-enabled automated drip irrigation system with real-time water-level monitoring and adaptive control. The prototype integrated an ESP32 microcontroller with soil-moisture, temperature, humidity, and ultrasonic water-level sensors, a relay module, a water pump, and a drip irrigation arrangement. Sensor readings and pump status were transmitted to the Smart Irrigation IoT platform for real-time visualisation and remote monitoring. An adaptive threshold-based algorithm controlled pump operation according to soil-moisture conditions while preventing operation when the storage-tank water level was below the minimum threshold. The system was evaluated under controlled prototype conditions using sensing accuracy, response time, and water-saving performance. The recorded soil-moisture, temperature, and humidity accuracies were 97%, 95%, and 94%, respectively. The system achieved 40% water savings relative to conventional manual irrigation under identical test conditions and responded to threshold changes within 2 seconds. The prototype maintained communication with the IoT platform and supported automatic irrigation, water-level protection, remote access, and alert notifications. These findings indicate that the proposed system can support efficient irrigation management at the prototype scale. Long-term field trials across different crops, soil types, and climatic conditions are required to establish its reliability and scalability under practical agricultural conditions.

Explainable AI for Malaria Diagnosis: Comparative Analysis of ML Models Using Random Forest Feature Selection and SHAP Interpretability

David Kipngetich Chepkonga — 2026-07-31

Malaria remains one of the most significant causes of morbidity and mortality in tropical and subtropical regions, and timely diagnosis is essential for effective case management. Microscopy is the traditional parasitological reference standard, while rapid diagnostic tests (RDTs) are widely used, field-deployable alternatives with product- and antigen-dependent sensitivity and specificity; both approaches are constrained by requirements for trained personnel, reagents, or equipment in resource-limited settings. This study develops and evaluates, on a simulated clinical dataset calibrated to published aggregate statistics, an explainable artificial intelligence pipeline for malaria diagnosis prediction from routinely collectable symptoms, vital signs, and haematological indices, using six machine-learning (ML) models: Logistic Regression (LR), Naive Bayes (NB), K-Nearest Neighbours (KNN), Random Forest (RF), Support Vector Classifier (SVC), and Decision Tree (DT). The Synthetic Minority Oversampling Technique (SMOTE) and Random Forest feature selection are embedded within a single leakage-safe pipeline that is refitted in every cross-validation fold. The reported best model is selected on the basis of the cross-validated F1-score rather than held-out test performance, and the test set is used exactly once for confirmatory reporting. Under this design, SVC with RF-selected features was selected (mean cross-validated F1 = 0.740), achieving a test-set accuracy of 0.906 [95% CI 0.852, 0.953], recall of 0.867 [0.667, 1.000], and ROC AUC of 0.959 [0.919, 0.988]. A paired bootstrap test found no statistically significant difference in AUC compared with the runner-up, Logistic Regression (AUC 0.956, p = 0.81). Permutation importance corroborated 8 of the top 10 impurity-based features. Parasite density, the quantity used to determine the parasitological diagnosis, was excluded from the predictor set as a precautionary safeguard against near-total label leakage. Calibration, subgroup recall by age and sex, and a class-weighting comparison are also reported. This framework illustrates, without any claim of clinical validity, how a leakage-safe ML pipeline and SHAP interpretability can be combined and rigorously self-audited; real patient-level data and external validation are required before any clinical inference is drawn.