TA-FaultNet: A Temporal Attention Framework with Bidirectional LSTM for Multi-Class Fault Detection and Health Monitoring in Industrial Wireless Sensor Networks

Massila; Mustafa

doi:https://doi.org/10.54216/IJWAC.100105

International Journal of Artificial Intelligence and Education Technology | 2025

Public Education Investment, Instructional Resources, and Student Achievement: Cross-National Evidence from PISA 2022 in the Context of Sustainable Development Goal 4

All people must have access to educational opportunities which meet their needs through Sustainable Development Goal 4. The goal requires education systems to obtain sufficient funds and use their resources properly. The relationship between public education spending and student academic performance remains disputed because different countries achieve different results from their spending levels. The study employs PISA 2022 country-level scores which represent the first international assessment data published after COVID-19 to analyze public education expenditure as a GDP share together with pupil–teacher ratio and per-capita GDP in relation to student academic performance across three subjects. The study found that public education funding as percentage of GDP does not connect with PISA score results across 35 countries, showing no statistical link to tests (r = −0.095, p = 0.586). The pupil–teacher ratio serves as an effective predictor because it shows a strong negative relationship to student performance (β&circ; = −4.097, R2 = 0.312, p < 0.001). A three-variable regression model which combines expenditure share with pupil–teacher ratio and GDP per capita explains 59% of cross-country score variance (R2 = 0.592). High-income economies dominate the upper achievement tier, but several upper-middle-income systems— notably Estonia and Poland—substantially outperform their GDP-predicted  scores. The results show that organizations should focus their resources on developing teaching skills.

groups

Reshma Shaik mail -

Hanadi Osman Diab mail

link https://doi.org/10.54216/IJAIET.040203

Volume & Issue

Vol. Volume 4 / Iss. Issue 2

Details open_in_new

International Journal of BIM and Engineering Science | 2026

BIM-Integrated Semantic Risk Intelligence for Construction Safety Severity Prediction Using Incident Narratives and 4D Work-Zone Attributes

Construction safety management increasingly depends on the ability to connect static building information models with dynamic evidence from site operations. This paper proposes a BIM-integrated semantic risk intelligence model that translates accident narratives into work-zone risk indicators and uses them to infer safety severity. The model links textual incident evidence with BIM-relevant descriptors, including construction phase, spatial zone, temporary protection status, energy isolation, and proximity to safety constraints. A formal risk-scoring layer is combined with supervised severity learning to provide interpretable decision support for safety planning and 4D coordination. The study contributes a reproducible methodology for converting unstructured safety reports into BIM-actionable risk representations, supporting early hazard prioritisation, design-for-safety review, and site control planning. The findings indicate that semantic evidence becomes more useful when it is explicitly fused with BIM phase and spatial context, rather than being treated as disconnected textual data.

groups

Esam El-Mekawy mail

link https://doi.org/10.54216/IJBES.120205

Volume & Issue

Vol. Volume 12 / Iss. Issue 2

Details open_in_new

International Journal of BIM and Engineering Science | 2026

BIM Integration Across Engineering Disciplines: A Systematic Review of Methodological Advances, Interoperability Challenges, and Emerging Digital Frameworks

This paper provides a comprehensive systematic review of Building Information Modeling (BIM) integration across ten engineering disciplines, synthesising publications from January 2020 to January 2026. It identifies convergent trends, persistent knowledge gaps, and translational barriers that separate research prototypes from scalable industry practice. A PRISMA-guided systematic review was conducted across Scopus, Web of Science, ASCE Library, and ScienceDirect. An initial corpus of 4,712 records was screened and quality-assessed, yielding 63 papers for quantitative synthesis and a broader qualitative corpus of 293 studies spanning ten sub-domains: BIM–digital twin integration, BIM and artificial intelligence/machine learning, interoperability and IFC, structural engineering, MEP and building services, facility management and operations, BIM–GIS for smart cities, off-site and modular construction, adoption barriers, and energy and sustainability analysis. Annual BIM publications grew by approximately 256% between 2019 and 2024. BIM–AI/ML and BIM–digital twin integration are the two fastest-growing sub-domains, yet both remain constrained by data standardisation deficiencies and a shortage of domain-specific training datasets. IFC-based interoperability has matured significantly, but real-time bidirectional exchange across disciplines remains nascent. Structural engineering applications exhibit the highest technology readiness, while BIM–GIS integration for smart-city applications shows the widest gap between published prototypes and commercial deployment. The review delivers a thematic roadmap and a consolidated evidence base for prioritizing investment in digital workflows, standards development, and workforce training. An original four-layer integrated framework is proposed that connects engineering code provisions, AI/ML analytics, digital twin synchronisation, and automated quantity extraction within a single traceable workflow.

groups

Ann Wolter mail -

Paul Bailey mail -

Raja Ahmed Hassan mail -

Wipitha Mazungwi mail

link https://doi.org/10.54216/IJBES.120206

Volume & Issue

Vol. Volume 12 / Iss. Issue 2

Details open_in_new

International Journal of Artificial Intelligence and Education Technology | 2026

A Systematic Review of AI-Powered Uzbek Short-Answer Grading Using NLP and Teacher-Annotated Datasets

This paper presents a Systematic Literature Review (SLR) of AI-powered automated short-answer grading, with a particular focus on low-resource languages such as Uzbek. The review follows the PRISMA 2020 guidelines to ensure transparency and methodological rigor. Relevant peer-reviewed studies published between 2018 and 2025 were systematically identified, screened, and analyzed across multiple academic databases. In total, 33 studies were included in the final synthesis. The reviewed literature indicates that transformer-based models, including mBERT and XLM-R, generally achieve stronger performance than traditional machine learning approaches, while recent large language models show potential in few-shot and zero-shot grading scenarios. The findings also highlight that the limited availability of teacher-annotated datasets remains a major challenge for developing reliable automated grading systems in low-resource educational contexts.

groups

Sanjar Raximjonov mail -

Eugene Q. Castro mail

link https://doi.org/10.54216/IJAIET.050104

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

International Journal of Artificial Intelligence and Education Technology | 2026

Evaluating Microsoft Teams, Blackboard, Canvas, and Zoom for Online Teaching Effectiveness: A Multi-Dimensional Comparative Study in Higher Education

The rapid institutionalisation of online and hybrid delivery models in higher education has left instructors and academic administrators managing a fragmented landscape of dedicated learning management systems, video conferencing platforms, and collaborative productivity suites that overlap substantially in function but differ markedly in pedagogical affordance. Selecting a platform or combination of platforms is consequential for instructor workload, student engagement, and learning outcomes, yet the evidence base for such decisions remains limited to narrow singleplatform evaluations or anecdotal comparisons. This paper presents a systematic multi-dimensional comparative evaluation of four widely adopted platforms—Microsoft Teams, Blackboard, Canvas, and Zoom—drawing on original survey data from 284 instructors and 642 students across five higher education institutions. Nine evaluation dimensions are examined: content delivery, real-time collaboration, assessment and feedback, usability, technical reliability, student engagement support, accessibility, analytics and reporting, and third-party integration. Quantitative analyses include one-way analysis of variance across all nine dimensions, Bonferroni post-hoc comparisons, Pearson correlation analysis, and multiple regression modelling of the predictors of instructor overall satisfaction. Canvas achieves the highest composite scores for usability, analytics, and integration; Blackboard leads on assessment and reporting depth; Microsoft Teams leads on real-time collaboration; and Zoom leads on content delivery in synchronous sessions but performs poorly on the asynchronous dimensions where dedicated learning management systems are strongest. The paper synthesizes findings into a platform selection framework and eight evidence-based recommendations for practitioners designing or evaluating technology-enhanced teaching environments.

groups

Tariq Saali mail -

Tassawar Kamran mail

link https://doi.org/10.54216/IJAIET.050105

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

International Journal of Artificial Intelligence and Education Technology | 2026

ChatGPT as an Assessment Design Tool in Higher Education: Evaluating Item Quality, Bloom’s Taxonomy Coverage, and Faculty Acceptance Across Academic Disciplines

The emergence of large language models capable of generating coherent, contextually grounded text at scale has created a new and contested tool for higher education assessment design: instructors can now produce examination questions, assignment prompts, and feedback rubrics in seconds rather than hours. Whether the items produced by these systems meet the quality standards required for valid, reliable, and pedagogically appropriate higher education assessment is an empirical question that the literature has only partially addressed. This paper reports a three-study investigation of ChatGPT as an assessment design tool in higher education, covering item quality, cognitive level coverage, student performance, and faculty acceptance. Study 1 presents an expert-panel evaluation of 360 assessment items—180 generated by ChatGPT and 180 created by experienced instructors across six academic disciplines and four item types, rated on seven quality dimensions including content accuracy, Bloom’s taxonomy alignment, linguistic clarity, and originality. Study 2 reports a faculty survey of 186 instructors examining adoption rates, perceived benefits, concerns, and the predictors of acceptance. Study 3 compares the performance of 412 students on counterbalanced ChatGPT-generated and instructor-created assessment items. ChatGPT-generated items score significantly below instructor-created items on Bloom’s taxonomy alignment and originality, but perform comparably or above on linguistic clarity and difficulty calibration. Student performance is modestly but significantly higher on ChatGPT-generated items, a finding that challenges simple assumptions about AI-generated assessment difficulty. Academic integrity concerns and higher-order cognitive coverage are the dominant faculty concerns, while time savings—averaging 77% reduction in item-writing time—is the most consistently cited benefit. The paper contributes a validated multi-dimensional item quality framework, a faculty acceptance model, and eight evidence-based guidelines for the responsible integration of ChatGPT in assessment design workflows.

groups

Nadia Iftikhar mail -

Rabia Muslu mail

link https://doi.org/10.54216/IJAIET.050106

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

International Journal of Wireless and Ad Hoc Communication | 2026

Communication-A ware Digital-Twin Reliability Budgeting for Fog-Assisted Wireless Sensor Ad Hoc Networks

Wireless sensor IoT systems are increasingly deployed as infrastructure-light communication fabrics in which battery-powered devices exchange event streams through local gateways, fog nodes, and sometimes multi-hop ad hoc routes. In such settings, reliability cannot be judged only by how fast a packet reaches a server. A reading may be fresh but untrusted, energy-efficient but delayed, or successfully delivered through a route that overloads the next fog node. This article revises the problem as a communication-aware reliability budgeting task for fog assisted wireless sensor ad hoc networks. It reviews core studies on wireless sensor networking, fog and edge computing, digital twins, edge intelligence, federated learning, and IoT security, then introduces an extended Digital-Twin Reliability Budgeting model. The model maintains compact fog-side twin states and uses them to govern route choice, event compression, fog offloading, replication, and cloud escalation. Three mathematical algorithms are presented for twin synchronization, route-and-action selection, and adaptive budget learning. The analysis develops delay, energy, freshness, loss, trust, and occupancy terms and shows how they interact across multi-hop communication paths. The resulting framework supports a more disciplined design philosophy: fog nodes should not only process sensor data near the edge; they should regulate the reliability budget of each communication decision before network resources are consumed.

groups

Safina Shokeen mail -

Vishal Srivastava mail

link https://doi.org/10.54216/IJWAC.100106

Volume & Issue

Vol. Volume 10 / Iss. Issue 1

Details open_in_new

International Journal of Wireless and Ad Hoc Communication | 2026

Edge-Bandwidth Brokerage for RFID-Enabled Ad Hoc IoT Communication Networks

RFID-enabled IoT deployments often lose communication efficiency not because a single tag message is large, but because tag observations are repeated, partially redundant, and reported by overlapping readers. In an edgecomputing environment, this redundancy becomes a bandwidth-governance problem: the local gateway must decide what is worth forwarding, what can be compressed, and what should be kept only as short-term local evidence. This article presents an edge-bandwidth brokerage model for RFID-assisted ad hoc IoT communication networks. The proposed model, named BASER, interprets every tag read as a priced communication event whose forwarding value depends on novelty, duplication risk, priority, motion context, and instantaneous backhaul pressure. The paper develops a three-stage mathematical formulation for value construction, budgeted admission, and adaptive compression. A reproducible scenario analysis is used to study how tag density, mobility, edge load, and uplink capacity affect latency, loss, semantic retention, and energy consumption. The main finding is that bandwidth savings should not be treated as a blind compression target; instead, the edge node should act as a broker that protects meaningful RFID events while preventing repeated low-value reads from saturating the uplink.

groups

Salah-ddine KRIT mail

link https://doi.org/10.54216/IJWAC.100207

Volume & Issue

Vol. Volume 10 / Iss. Issue 2

Details open_in_new

Journal of Cybersecurity and Information Management | 2026

Machine Learning-Driven Cyber Threat Prediction and Prevention: A Multi-Dataset Design and Comparative Evaluation

As technology advances, the frequency and variety of intrusions and other security threats within network environments continue to grow. Intrusion detection systems (IDS) play a vital role in securing networks against unauthorized access and attacks on computer systems; however, traditional IDSs are very limited in their ability to recognize new, complex malicious threats because they rely on signature-based detection. Approaches based on machine learning have shown a promising alternative in identifying unknown malicious attacks. This study proposes a computationally efficient, generalizable machine-learning framework for robust cyber-threat prediction. Three benchmark datasets (HIKARI-2021, CIC-IDS2017, and KDDCup99) were used for full-pipeline evaluations, including preprocessing, feature selection, class-imbalance handling, hyperparameter optimization, and strict model validation. Eight classifiers were assessed, which included traditional classifiers and more modern ensemble methods. The results from this study showed that tree-based models, mainly both Random Forest and XGBoost achieved near-perfect performance across all datasets, reaching accuracy values up to 0.999 and F1-scores between 0.99 and 0.999. Additionally, the SHAP-based explainability analysis was applied to reveal features that drove predictions, enabling interpretability and transparency. Compared with prior studies, the proposed framework consistently delivers improved, more stable detection performance. The findings highlight that optimized ML models combined with balanced datasets and rigorous validation protocols can significantly enhance intrusion detection reliability. Furthermore, this approach provides a practical and scalable solution for strengthening cybersecurity defenses against evolving and emerging cyber threats.

groups

Krishneel Sundar mail -

Pritika Reddy mail -

Kaylash C. Chaudhary mail

link https://doi.org/10.54216/JCIM.180106

Volume & Issue

Vol. Volume 18 / Iss. Issue 1

Details open_in_new

International Journal of Wireless and Ad Hoc Communication | 2026

TA-FaultNet: A Temporal Attention Framework with Bidirectional LSTM for Multi-Class Fault Detection and Health Monitoring in Industrial Wireless Sensor Networks

Industrial wireless sensor networks are central to the continuous monitoring of critical plant equipment, yet reliable identification of multiple concurrent fault modes from heterogeneous multivariate sensor streams remains an unsolved operational challenge. Physical failure mechanisms—pump cavitation, valve blockage, gradual sensor drift—and wireless channel disturbances each imprint distinct but overlapping temporal signatures that render classical thresholdand rule-based detectors inadequate for automated maintenance dispatch. This paper  presents TA-FaultNet, a neural architecture designed specifically for the multi-class fault identification problem in industrial sensor deployments. The network couples a two-stage stacked bidirectional recurrent encoder with a parallel multi-head self-attention module and a compact temporal convolutional block, enabling simultaneous capture of long-range process dynamics and fine-grained fault-onset localisation from raw sensor windows. TA-FaultNet is evaluated on the publicly available Skoltech Anomaly Benchmark under five operational classes and assessed through a comprehensive battery of experiments including baseline comparisons, systematic component ablation, cross-experiment generalisation, andprogressive noise-injection testing. The proposed architecture decisively outperforms eight competing methods spanning classical anomaly detectors, standalone recurrent and convolutional networks, and the Transformer, while remaining lightweight enough for edge gateway deployment. Attention weight visualisations expose fault-specific temporal activation patterns, providing maintenance engineers with interpretable diagnostic evidence beyond bare classification labels.

groups

Massila Kamalrudin mail -

Mustafa Musa mail

link https://doi.org/10.54216/IJWAC.100105

Volume & Issue

Vol. Volume 10 / Iss. Issue 1

Details open_in_new