Journal of Information and Organizational Sciences (JIOS)
Not a member yet
455 research outputs found
Sort by
Comparative Study of VGG16, ResNet50, and YOLOv8 Models in Detecting Driver Distraction in Varying Lighting Conditions
No full textObserving driver distractions while driving gives valuable information to prevent accidents, so it is necessary to use effective monitoring methods. Deep learning is showing new capabilities in solving this issue. This study evaluates the results of CNN, YOLOv8, ResNet50 and VGG16 deep learning models as they detect drivers who are practising distracted driving behaviours under real-time and various lighting conditions (day and night). The models were trained on two datasets: the labelled State Farm dataset and the Driver Monitor Dataset (DMD). They successfully identified ten distinct categories of distraction for the State Farm dataset and five categories for the monitoring drivers dataset. Pre-trained models were optimized using transfer learning through fine-tuning to enhance detection accuracy. This paper studies related work on distracted driving and shares ideas for designing advanced systems that use various methods to improve accuracy. YOLOv8 reached an outstanding test accuracy of 98.46% on the State Farm dataset, proving itself superior to other methods and confirming its effectiveness for monitoring. In addition, YOLOv8 reached 96.46% accuracy in the DMD dataset, outperforming VGG16 at 90.58% and ResNet50 at 70.80%. YOLOv8 was able to recognise important driver behaviours in real time with a dataset of 15 subjects and 20 different driving postures. The research proves that the YOLOv8 model is fit for use in intelligent monitoring systems designed to detect distracted driving and promote safer driving through focused actions
A Hybrid IG-PCA and Machine Learning Approach for Accurate Intrusion Detection in IoMT with Imbalanced Data
No full textThe rapid growth of the Internet of Medical Things (IoMT) has introduced critical cybersecurity challenges, highlighting the need for robust and accurate intrusion detection systems (IDS). This study presents a hybrid machine learning (ML) framework to strengthen intrusion detection in IoMT networks using the CIC-IoMT2024 dataset. The framework combines Information Gain (IG) and Principal Component Analysis (PCA) for feature selection and dimensionality reduction, while SMOTEENN and SMOTETomek are applied to address severe class imbalance. The processed data are classified using Random Forest (RF), K-Nearest Neighbors (KNN), XGBoost (XGB), Multi-Layer Perceptron (MLPC), and Logistic Regression (LR), with hyperparameters optimized through Bayesian Optimization. Performance is evaluated using Accuracy, Precision, Recall, F1-Score, and AUC. Experimental results reveal that the optimized XGB classifier with SMOTEENN achieves a peak accuracy of 99.811%. This top-tier performance surpasses several existing benchmarks, validating the effectiveness of integrating IG-PCA with advanced resampling and optimization strategies. This work contributes a lightweight, scalable, and highly accurate IDS, offering a practical and efficient solution for enhancing security in resource-constrained, next-generation medical IoT systems
Model Checking Access Control Protocol for Spreadsheets
No full textSpreadsheets are one of the most used software systems in business and academia. Since the first introduction of electronic spreadsheets for personal computers in 1979, spreadsheets have significantly evolved. With recent technological advancements and new features added, spreadsheets have become powerful computing platforms capable of complex analysis and modelling. However, numerous publications over the years described cases of spreadsheet errors. In focus of this research paper are spreadsheet errors caused by unauthorized access and modifications of spreadsheets in multi-user environments. Specifically, this paper is structured around formal verification of the novel ABAC4S (Attribute Based Access Control for Spreadsheets) protocol designed for prevention or detection of unauthorized modifications to spreadsheets in multi-user environments. We utilized a model checking approach to verify ABAC4S protocol rules for correctness
Developing a Shared Knowledge Area Mechanism for Multi-Mobile Agents to Improve Performance Using Machine Learning: Classification-Rule
No full textA mobile agent system is a mobile computing approach where agents move autonomously among hosts to perform tasks. It offers advantages such as low latency, reduced bandwidth use, and cost efficiency. This paper proposes the Shared Knowledge Area Mechanism (SKAM) to improve mobile agent performance. SKAM uses a shared knowledge database that stores classification rules based on agents’ travel experiences. Each rule is an IF–THEN statement linking service combinations to host locations. We extract these rules using support, confidence, and lift to ensure reliability. Before starting a task, an agent queries the database to select hosts based on the most relevant rules. This reduces unnecessary host visits and shortens travel time. SKAM is implemented within the Secure Mobile Agent Generator (SMAG), a platform used to simulate mobile agent behavior. SKAM also applies rule prioritization to support accurate itinerary planning. Experimental results show that SKAM reduces average task completion time from 41,146.5 ms to 23,445.5 ms—a 43% improvement. This gain is statistically significant (p < 0.05) and consistent across all agents. It confirms that SKAM lowers both search overhead and travel time. These results highlight SKAM’s effectiveness and practical value for real-time, large-scale mobile agent systems
Towards the Institutional Student Mobility Ecosystem (ISME): model based on systematic literature review
No full textThe internationalization of higher education, especially in terms of student mobility, is a key indicator of the quality of higher education. Additionally, increasing the number of students participating in credit mobility is one of the strategic goals of the European higher education area. To help more students take advantage of student mobility programs, it is essential to understand the factors that influence student mobility at both the institutional and individual levels. This paper proposes the Institutional Student Mobility Ecosystem (ISME) for credit mobility. It is based on a systematic literature review of 321 initially retrieved sources, with 22 analyzed in detail. The results are supplemented by the analysis of 11 policy and professional documents. The proposed ISME identifies student decisional factors, supporting mechanisms and stakeholders as key enablers of student mobility. Additionally, it outlines the outcomes for students, HEIs and society that result from student mobility. This model provides valuable groundwork for researchers in the field of student mobility, facilitating further in-depth analysis of specific elements within the student mobility ecosystem
Assessing User Satisfaction of Local Government Websites Through ISO 25010 and Technology Acceptance Model (TAM): A SmartPLS and IPMA-Based Study in Lombok Tengah
No full textDespite continued efforts to digitize public services, many local government websites in emerging contexts still underperform in delivering satisfactory user experiences. This study develops an integrated evaluation framework that combines the ISO 25010 software quality model with the Technology Acceptance Model (TAM) to jointly assess system quality and user acceptance. We analyzed survey data from 524 users in Lombok Tengah, Indonesia, using Partial Least Squares Structural Equation Modeling (PLS-SEM) and Importance Performance Map Analysis (IPMA). The results indicate that functional suitability, usability, and reliability significantly shape perceived usefulness, whereas reliability, security, and performance efficiency drive perceived ease of use. Both perceived usefulness and perceived ease of use positively influence user satisfaction and behavioral intention, with satisfaction emerging as the strongest predictor. IPMA highlights performance efficiency and security as priority areas for improvement. The study contributes to e-government literature by proposing a dual layer model that links system level attributes to user-level perceptions and outcomes, and by translating statistical effects into actionable priorities for local governments seeking to enhance the quality and adoption of digital public services in semi urban developing regions
Evaluating The Flipped Classroom Approach in Computer Science Curricula
No full textActive, technology-supported learning accelerated during and after COVID-19, yet evidence from non-programming computer science courses remains limited. This paper contributes (i) a focused review of flipped classroom (FC) studies in CS program (2020-2024) and (ii) a three-year case study of how the flipped classroom enhances the teaching of IT Service Management (ITSM) as a discipline in the computer science program in an online university environment, during and after the COVID-19 pandemic. The FC design combined pre-lecture micro-videos and auto graded quizzes with in-classroom clarification and post classroom activities (project).
Using LMS telemetry, course outcomes, and an end of semester survey across three academic years (2021/22-2023/24), we examined engagement-achievement links with non-parametric, rank based correlations (Spearman ρ), regularized logistic regression, and comparisons across empirically defined engagement tertiles. Results show consistent, practically meaningful associations between quality weighted engagement (quiz participation and performance) and both passing and final grades, with survey perceptions aligning to the behavioral signals. While strictly non-causal, the pattern is robust across methods and suggests actionable uses: early identification of at-risk students and design guidance that emphasizes short, well scaffolded videos and steady formative assessment
Optimized Offloading in Vehicular Edge Computing: A Game Theoretic Analysis
No full textThis paper introduces a novel approach to Vehicle Edge Computing (VEC), addressing the need for low-latency, high-reliability applications in vehicle networks. By leveraging nearby Multi-access Edge Computing (MEC) resources, VEC enhances data processing speed and reliability for applications like autonomous driving, real-time traffic management, and infotainment systems. The proposed solution models a multi-user non-cooperative computation offloading game in vehicular MEC networks, where each vehicle adjusts its offloading probability based on factors like distance to the MEC access point, communication model, and competition for resources. Additionally, a best response-learning algorithm is designed based on the computation offloading game model. The approach focuses on maximizing each vehicle’s utility while ensuring convergence to a single, stable equilibrium under defined conditions. To demonstrate the effectiveness and performance of the proposed algorithms, comprehensive experiments were performed. Numerical results demonstrate the fast convergence and improved performance achieved
Hierarchical Deep Learning Model Optimization Using Enhanced Evolutionary-based Approach for Fake News Detection
No full textMultimodal fake information on social media is a growing concern worldwide. Existing deep learning-based solutions typically involve designing hierarchical models that capture relevant features from each modality, which are then fused for final classification. However, these models are often complex, with numerous trainable parameters, making them resource-intensive. This work introduces the Deep Learning Model with Evolutionary Computing Approach (DLECA), a novel method for compressing and optimizing hierarchical deep learning models (HDLM). It employs an enhanced genetic algorithm (GA) with a unique fitness function, dynamic crossover, and adaptive mutation strategies to achieve model compression, maintain accuracy, and balance exploration and exploitation during evolution. In comparison to manually designed HDLM, the proposed approach achieves up to 97.86% model compression with a 0.34% accuracy improvement, while a variant achieves 96.24% compression with a 0.23% accuracy improvement. Comparative analysis shows that DLECA outperforms Random Walk and Bayesian Optimization in multimodal fake news detection, offering a more efficient and accurate solution