Journal of Transactions in Systems Engineering

A Systematic Literature Review on Integrating VANETs, VDTNs, 5G, and IoT for Smart Cities: Current Approaches, Challenges, and Future Directions

2025-09-21T18:03:38+02:00

Nowadays smart cities have become a necessity for rapidly changing and transforming urban environments and the core technologies enabling this development are Vehicular Ad Hoc Networks (VANETs), Vehicular Delay Tolerant Networks (VDTNs), 5G networks, and Internet of Things (IoT). These technologies alone offer an important contribution, but when integrated effectively, they offer the opportunity of uninterrupted connectivity, real-time data sharing and management of urban resources. This paper conducts a comprehensive literature review to study existing techniques/approaches and challenges for integrating VANET, VDTN, 5G, and IoT within smart cities based on three research questions. Recent articles from databases such as Google Scholar, ResearchGate and MDPI, were reviewed to examine this integration, to identify recent advancements in this topic with focus on innovative methodologies proposed in an international context and to highlight the research gaps, challenges and solutions. The VOSviewer software was used to build the keyword co-occurrence network and to cluster the relevant literature. Our findings reveal that although promising solutions exist, issues such as high mobility, heterogeneous network architecture, and resource constraints remain critical barriers to large-scale deployment of smart city applications. Furthermore, this review proposes a conceptual framework for intelligent and adaptive network integration of VANET, VDTN, IoT, and 5G for future smart city applications.

AI-Enabled Distributed Cloud Frameworks for Big Data Analytics with Privacy Preservation

2025-11-18T05:46:23+01:00

The fast rise of the Big Data and the faster adoption of Artificial Intelligence (AI) have changed the modern computational ecosystems allowing to conduct real-time analytics and automation in some of the most vital areas: healthcare, finance, and industrial IoT. Nevertheless, there are notable problems with traditional centralized cloud designs such as poor scalability, high latency, network overload, and augmented privacy and security threats in distributing sensitive information that is sensitive. To fill these gaps, this paper suggests a new AI-enabled distributed cloud framework (AIDCF) that combines federated learning, differential privacy, and homomorphic encryption to facilitate secure, privacy-preserving and scalable analytics on the Big Data without centralized sharing of data. The mixed-method research design was chosen, which involved the development of theoretical frameworks, the modeling of algorithms, and simulation of experiment, based on synthetic multi-domain data (healthcare, finance, IoT) running on distributed cloud nodes (10-100). The outcomes indicate the novelty and high-performance AIDCF, which has a high accuracy (93.7%), low latency (139 ms), high throughput (1585 MB/s) and high computational performance (89.5 percent), as well as the significant reduction in privacy loss (ε = 1.3) compared to other models such as DP-FedAvg, SecureML, and Baseline Cloud Analytics (BCA). These results confirm that the presented framework provides a feasible trade-off between analytical and confidentiality protection, which makes it be deployed in privacy-sensitive, real-time, and large-scale distributed systems. Altogether, AIDCF offers a scalable, secure, and high-performance distributed AI system, which will push the state of the art of privacy-aware Big Data processing.

Machine Learning Analysis of Social Media Usage Patterns and Mental Health Indicators

2025-12-18T21:32:52+01:00

The rapid growth of social media has created new opportunities for connection but has also raised concerns about its impact on mental health. This study investigates how demographic factors, digital behaviour, and self-reported psychological indicators jointly relate to users’ mental states. Using an open dataset of 5000 social media users, we analyse numerical and categorical variables including age, gender, daily screen time, social media time, counts of positive and negative interactions, sleep duration, physical activity, anxiety, stress, mood, and a three-level mental-state label (Healthy, At_Risk, Stressed). Descriptive statistics and correlation analysis show that longer daily screen and social media time are strongly associated with higher stress and anxiety and lower mood, while sleep and physical activity display the opposite pattern. K-means clustering applied to combined behavioural and psychological features reveals three coherent user profiles that align with the Healthy, At_Risk, and Stressed categories, highlighting a clear gradient from balanced to high-risk digital lifestyles. A decision-tree classifier trained only on behavioural features (excluding anxiety, stress, and mood to avoid target leakage) achieves an overall accuracy of about 97% on a held-out test set and provides interpretable if then rules linking specific usage patterns to mental states. The results emphasise that intensive, unbalanced social media use especially when coupled with reduced sleep and low physical activity is strongly linked to adverse mental-health outcomes, and they illustrate how simple machine-learning models can support early risk detection based on non-intrusive behavioural data.