×
The submission system is temporarily under maintenance. Please send your manuscripts to
Go to Editorial ManagerThis narrative study provides an analytical and critical review of recent advancements (2019-2026) in the integration of IoT (Internet of Things) and AI (Artificial Intelligence) systems for fall detection and child health monitoring. Unlike prior studies, which concentrated on elderly care and monitoring, this study examines child-specific monitoring environments, including wearable, vision-based, and hybrid systems. It investigates new trends such as the combination of deep learning and interpretable AI with multimedia sensory input and peripheral or fuzzy computing. Data scarcity, real-world deployment limits, privacy concerns, and age-related changes are among the key challenges addressed. The paper identifies important research gaps and proposes future paths for sustainable, secure, and accessible intelligent child monitoring systems.
Authenticity of tests as a measurement tool has received a lot of attention within learning institutions due to emergences of online classes and remote test administration. Supervision and invigilation methods do not always suffice to deter students from cheating, and thus Academic Cheating Detection Systems (ACDETS) have been invented. This paper presents a critical analysis of the current approaches for identifying cheating in online and face-to-face examination systems. There are plenty of approaches, including behavioral approach, facial expressions tracking, gestures recognition, voice analysis, and video monitoring. CNN (Convolutional Neural Network) algorithms, RNN (Recurrent Neural Network) algorithms, and YOLO models, for instance, have shown great enhancements in both accuracy and scalability of detecting suspicious behaviors. The paper further compares the merits and demerits of these methods and also looks at the possibility of using them for real time detection, large setting for exams, and varied testing conditions. This paper is finalized by the evaluation of the practical applicability of the findings, limitations, and further research prospects concerning the monitoring of academic integrity.
This study analyzes homicide data in the United States from 1980 to 2014 using machine learning techniques to predict crime resolution and classify victim gender. The dataset, obtained from the FBI Supplementary Homicide Report, contains 638,454 records. Data preprocessing involved cleaning, converting categorical features to numerical values, and addressing class imbalance using Synthetic Minority Oversampling Technique (SMOTE). Various classification algorithms were applied, including Decision Tree and Naïve Bayes. The results showed that the Decision Tree model achieved 95% accuracy in predicting crime resolution and 85% accuracy in classifying victim gender, while Naïve Bayes reached 92% accuracy in crime resolution prediction. The findings highlight the effectiveness of machine learning in crime pattern analysis and prediction, aiding law enforcement in making more informed investigative decisions.
The combination of the Internet of Things (IoT) and Software-Defined Networking (SDN) has quickly revolutionized the world of industry, and also brought with it some severe security risks. Despite its recent success, Deep learning (DL) can be a powerful method for Network Intrusion Detection systems (NIDS), but often the computational and memory requirements are too high for edge devices. In this paper, a collaborative optimization framework with lightweight deep-learning IDS models suitable for SDN-IoT environments is proposed. We formalize and test three collaborative compression schemes based on combining pruning, clustering, and quantization, yet maintaining the structural advantages of each stage: Sparsity-Preserving Quantization-Aware Training (PQAT), Cluster-Preserving Quantization-Aware Training (CQAT), and Sparsity-and-Cluster-Preserving Quantization-Aware Training (PCQAT). The three schemes are evaluated on four canonical architectures (LSTM, RNN, DNN and CNN) on the proposed SICA dataset with respect to DDoS, Probe and Brute-Force (BFA).The results show that aggressive single-stage compression such as clustering can shrink model size by up to 94%, but at the cost of predictive confidence; PCQAT, by contrast, acts as a stabilizer that recovers this lost confidence. In particular, CNN-PCQAT and DNN-PCQAT deliver the most favorable trade-off, sustaining high detection accuracy on all three attack classes with inference latency below 0.01 ms. The framework therefore offers a practical, deployment-oriented guideline for implementing real-time IDS on resource-constrained SDN-IoT edge devices.