Volume 7 Issue 12 2025

With the rapid development of the Internet, social networks have become one of the most widely used platforms. Users frequently share their moments, thoughts, and emotions through these channels. However, this surge in social media usage has also led to increased cybercrime in recent years. Cybercrime can manifest itself in various forms, including hate speech, bullying, scams, and more. Given the vastness of the Internet, manually collecting and analyzing such data is tedious, impractical, and time-consuming. This research proposes a cross-platform tool designed to automate the scraping and analysis of a suspected cybercriminal’s activity on multiple social media accounts. It should be noted that all data used was collected with proper permission taken . The data used was from our close friends only. The data is neither stored nor shared with any third party

The transformation of higher education in South Africa remains constrained by persistent inequalities in access, equity, and curriculum relevance, with rural communities facing the most acute barriers. Students from rural areas experience systemic exclusion through inadequate schooling, poor digital infrastructure, limited financial support and geographic marginalisation. The onset of the Fourth Industrial Revolution (4IR) has intensified these challenges by placing new demands on digital literacy, data skills and technological adaptability. While 4IR technologies present opportunities for rethinking curricula and expanding participation, they risk reinforcing rural exclusion if policies do not foreground equity. This paper examines the intersection of rural development, higher education, and the 4th Industrial Revolution (4IR) within South Africa’s 2025 Presidency of the Group of 20 (G20). It considers the rural–urban digital divide, evaluates curriculum transformation to balance African epistemologies with global digital skills, and positions rural development as a G20 priority through proposals such as a Rural Digital Compact. The paper further explores the role of universities as rural innovation hubs. It concludes with policy recommendations to embed bursaries, rural campuses, entrepreneurial curricula and community partnerships at the centre of transformation. South Africa can leverage its G20 Presidency to advance inclusive higher education and sustainable growth by foregrounding rural development.

This research addresses the challenge of multi-class mental disorder diagnosis (using behavioral, psychological metrics) while preserving the privacy of the data and providing post hoc explanations to a set of humans. Error-prone samples were removed using deep autoencoder based global anomaly detection technique that significantly improved data quality as reflected through multiple number of high-AUC machine learning models' performances over the 100,000 records datasets. For reliable mental disorder classification, the study uses an extensive multi-model framework that integrates MLP, SVM, Random Forest, LightGBM, CatBoost, XGBoost, KNN, Naive Bayes and a Stacking Ensemble. Out of all the models trained, MLP domains with the highest metrics 96.08% accuracy, 0.9608 macro-F1, MCC 0.9477 and Cohen’s Kappa 0.9477. Almost 5% of these patterns were revealed by feeding the data into an anomaly detection system. Proposed system revealed enviably high-risk metrics using SHAP. In order to create an accurate and intelligible evaluation system, this work has accomplished secure data preprocessing, anomaly filtering, multiclass mental disorder classification and SHAP explanation based on the SHAP. In addition to providing direct psychological risk exposure, the suggested integrated model ensures privacy protection, a high-quality single-cleaned sample and equitable predictive performance.

This paper presents an analytical study on efficient power flow control in grid-connected photovoltaic and battery systems, emphasizing their capability to enhance power quality and ensure stable grid interaction. A comprehensive model is developed in MATLAB/Simulink, integrating PV generation with battery storage and shunt compensation to regulate voltage, mitigate harmonics and maintain optimal power exchange with the grid. The PV array is operated under variable irradiance conditions while the battery compensates for fluctuations through controlled charging and discharging. A robust maximum power point tracking (MPPT) algorithm ensures rapid convergence of the PV operating point enabling effective utilization of solar energy. The battery–converter interface is analyzed for voltage stability and current dynamics during abrupt load changes. Shunt inverters are investigated for reactive power support and harmonic suppression contributing to enhanced voltage regulation at the point of common coupling. Simulation results confirm that the proposed control framework achieves efficient power balancing among the PV array, battery and utility grid even under transient disturbances. Grid voltage and current waveforms remain well-synchronized and load-side power quality is preserved despite nonlinear demand. The study demonstrates that coordinated operation of PV and battery resources, supported by advanced control of interfacing converters provides a resilient and efficient solution for integrating renewable energy into low-voltage distribution networks. The findings offer practical insights for designing smart grid systems capable of sustaining reliable power delivery while maximizing renewable energy penetration.

The availability of civic and political action is crucially limited under the occupation of the West Bank, forcing the Palestinian youth to base their alternative expression and organised activism on social media. The research problem of the proposed study is to determine whether cognitive, evaluative, and affective aspects of social identity affect Palestinian youth activation and whether social media mediates the relationship between social media and social identity. The West Bank was surveyed across 400 and sampled youths (18 to 29) in a cross-sectional quantitative survey. The resulting data were processed with the help of Partial Least Squares Structural Equation Modeling (PLS-SEM) to evaluate both direct and indirect links between the dimensions of social identity, use of social media, and the degree of activism. The findings demonstrate that the three dimensions of identity are important predictors of activism, and social media acts as a mediator between identity and participation to some extent. The strongest effect was observed in the affective dimension, implying that political participation by young people is largely influenced by emotional attachment. The model accounted for 67 per cent of the activism variance, which is high predictive power. The findings show that identity is mediated by social media in connection with collective action. The study helps comprehend better how the social identity becomes political action in the online space, and provides a holistic view of how the younger generations can use digital space to maintain agency, solidarity, and civic participation despite those limiting conditions.

In this paper we construct the models foliations on pseudo Anosov bundle using the Gyhs Thurston’s method which consists of the suspension of diffeomorphism followed by desingularization and we study some properties of those foliations.

In this paper we study the geometry of Mannheim curves in a strict Walker 3-manifold and we obtain explicit parametric equations for Mannheim curves and timelike Mannheim curves, respectively. We determine the distance between two corresponding points of the Mannheim pair of curves and show that distance depending of the curvature. We discuss the relationship between the curvature and torsion of a pair of Mannheim curves in a strict Walker Manifold. We finish by an example of Mannheim pair curves for illustrate the result.

In this article, we consider the case of a birth-death process (B.D.P.) with an infinite number of states and constant birth and death rates (independent of the number of states). We have determined the deterministic solution (the distribution of the B.D.P.) of the Kolmogorov differential equations for the birth-death process, a solution that depends only on the birth (lambda) and death (mu) rates, and this is a first for scientific research.

The prospects for a circular economy in the management of low-value plastic waste are highly promising, offering environmental solutions alongside substantial economic opportunities through innovative recycling into high-value products. Such an approach can transform challenges into economic opportunities, create employment, conserve natural resources, and reduce pollution through technological innovation. Mathematical modeling plays a crucial role by integrating data on costs, revenues, technologies, waste reduction, and environmental impacts, including emissions, thereby supporting strategic decision-making, scenario simulation, and investment justification to achieve economically viable and sustainable systems with minimal negative impacts. This study aims to analyze the prospects of a circular economy in processing lowvalue plastic waste into high-value products while minimizing environmental impacts using mathematical modeling approaches, including flow and mass balance models, economic models such as cost–benefit and profitability analysis, and environmental impact minimization models. The results provide quantitative estimates of material flows entering, being processed within, and exiting the plastic waste management system, measures of financial feasibility based on cost and revenue variables, and environmental impact metrics derived from life cycle assessment (LCA). Overall, this research contributes a robust quantitative framework for predicting the economic and environmental implications of circular plastic recycling and offers data-driven solutions to optimize waste-to-value processes while simultaneously reducing environmental pollution.

Plants diseases are one of the hardest challenges in world agriculture where it has traditionally been diagnosed by farmers or experts that perform manual inspection. This traditional method is slow, subjective and erratic, resulting in delays in treatment and considerable loss of crops. With the advent of artificial intelligence, new applications nowadays allow automated, data-driven solutions to optimize agricultural accuracy and efficiency. Here we report an ongoing problem of disease identification being too late and not always accurate due to relying on visual observation and having few experts for consultation. A solution to these challenges is considered in the proposed system, which utilizes an AI/MLbased model to identify plant disease right from leaf images. The procedure involves application of image pre-processing, feature extraction and deep learning classification algorithm using Convolutional Neural Networks(CNNs) to successfully detect disease types with higher accuracy. Moreover, the system is also equipped with a treatment-recommendation block that associates each determined disease to appropriate cures, providing farmers with an immediate actionable advice. The results indicate that the model can efficiently distinguish healthy leaves and ill ones across various plant species with good accuracy rate; enabling to give a prompt diagnosis and treatment if necessary. With this combination significant reduction of human dependency is achieved and the burden on the crop damage is also reduced, leading to sustainable technology-based farming.

In the contemporary knowledge-driven economy, knowledge management, organizational learning culture, and intellectual capital management have become critical foundations for organizational adaptation, innovation, and organizational performance. Although extensive research has examined these knowledge-based resources in business and industrial contexts, empirical studies in religious and non-profit educational institutions remain limited. This qualitative study aims to explore how knowledge management, organizational learning culture, and intellectual capital management are perceived and practiced, how innovation capability develops, and how these factors influence organizational performance in temple-based educational institutions. Data were collected through in-depth interviews and document analysis involving administrators and teachers and were analyzed using qualitative content analysis using a thematic approach. The findings indicate that intellectual capital management is predominantly informal and culturally embedded, relying heavily on shared values and religious traditions rather than formal systems. Human capital, particularly knowledge, experience, and moral values, emerged as the most salient dimension, while structural capital was reflected through routines, traditions, and religious regulations. Relational capital, especially trust-based relationships with communities and stakeholders, played a crucial role in sustaining organizational activities. Innovation capability was perceived as an incremental and adaptive process rooted in collective learning and knowledge sharing within a supportive organizational learning culture. Organizational performance was primarily understood in nonfinancial terms, emphasizing educational effectiveness, moral development, community satisfaction, and long-term sustainability. This study contributes to the literature by offering a contextualized understanding of knowledge-based resources and innovation in temple-based educational institutions.