Uplink and Downlink channel estimation in massive Multiple Input Multiple Output (MIMO) systems is an intricate issue because of the increasing channel matrix dimensions. The channel feedback overhead using traditional codebook schemes is very large, which consumes more bandwidth and decreases the overall system efficiency. The purpose of this paper is to decrease the channel estimation overhead by taking the advantage of sparse attributes and also to optimize the Energy Efficiency (EE) of the system. To cope with this issue, we propose a novel approach by using Compressed-Sensing (CS), Block Iterative-Support-Detection (Block-ISD), Angle-of-Departure (AoD) and Structured Compressive Sampling Matching Pursuit (S-CoSaMP) algorithms to reduce the channel estimation overhead and compare them with the traditional algorithms. The CS uses temporal-correlation of time-varying channels to produce Differential-Channel Impulse Response (DCIR) among two CIRs that are adjacent in time-slots. DCIR has greater sparsity than the conventional CIRs as it can be easily compressed. The Block-ISD uses spatial-correlation of the channels to obtain the block-sparsity which results in lower pilot-overhead. AoD quantizes the channels whose path-AoDs variation is slower than path-gains and such information is utilized for reducing the overhead. S-CoSaMP deploys structured-sparsity to obtain reliable Channel-State-Information (CSI). MATLAB simulation results show that the proposed CS based algorithms reduce the feedback and pilot-overhead by a significant percentage and also improve the system capacity as compared with the traditional algorithms. Moreover, the EE level increases with increasing Base Station (BS) density, UE density and lowering hardware impairments level.

The recent pandemic has raised significant challenges worldwide. In higher education, the necessity to adopt efficient strategies to sustain education during the crisis is mobilizing diverse, complementary, and integrative action in response. In this research article, we rise to the challenge of designing and implementing a transparent strategy for social media awareness at King Abdulaziz University (KAU). We introduce a framework for social media impact, termed the KAU Pandemic Framework. This includes the factors with the most important role in enhancing the deployment of social media in crisis in order to minimize the negative impact on education’s sustainability. We used a mixed-methods approach, integrating quantitative statistical analyses of social media data and online surveys and qualitative interviews in such a way as to construct a comprehensive framework. The results show that a methodological framework can be justified and that Twitter contributes significantly to six areas: administrative resilience; education sustainability; community responsibility; positive sentiment; community bonds; and delivery of promised value. The components of our proposed methodological framework integrate five pillars of the strategic adoption of social media: social media governance; social media resilience; social media utilization; decision-making capability; and institutional strategy. Finally, we show that the KAU Pandemic Framework can be used as strategic decision-making tool for the analysis of the gaps and inefficiencies in any social media plan that is deployed and the management challenges arising from the pandemic.

With the rapid growth of digital education, massive open online courses (MOOCs) have gained widespread popularity. However, high failure rates and low completion rates remain challenges, highlighting the need for accurate student performance prediction to identify at-risk students and implement timely interventions. Many existing models assume simplistic or linear dependencies, failing to capture the complex relationships and cross-dimensional interactions between different student features. Therefore, KANFormer model is developed, a novel deep learning (DL) architecture that integrates multi-head self attention (MHSA) mechanisms with Kolmogorov–Arnold networks (KAN) to effectively captures complex patterns within large-scale student datasets, enabling more accurate predictions of student performance. KANFormer processes student demographic, academic, and engagement features from the open university learning analytics dataset (OULAD) while applying a feature elimination process to enhance dataset quality, reduce redundancy, and mitigate multicollinearity, ensuring optimal predictive performance. The model is evaluated on both binary and multi-class classification tasks, distinguishing successful students (“Distinction” & “Pass”) from unsuccessful ones (“Fail” & “Withdrawn”), while also predicting individual performance categories. Ablation experiments analyze the impact of MHSA and KAN on model performance. Experimental results on the OULA dataset, comprising over 29,000 samples, demonstrate that KANFormer outperforms traditional and state-of-the-art models, achieving superior accuracy, precision, and F1-score in student performance prediction. To enhance interpretability, SHAP (SHapley Additive exPlanations) values are utilized to quantify each feature’s contribution to predictions. SHAP analysis shows that academic assessments (exam scores, TMAs, CMAs) are the most influential predictors of student success, while engagement indicators play a secondary role. Furthermore, the model’s generalization ability is validated on additional datasets, including ASSISTments2012, xAPI, and the Mathematics and Portuguese Language Courses dataset, proving its scalability and robustness across different educational contexts. By accurately identifying at-risk students, KANFormer offers a scalable, interpretable, and highly effective solution for improving student retention and success in online education, providing actionable insights for personalized learning interventions.

Realization of navigation in virtual environments remains a challenge as it involves complex operating conditions. Decomposition of such complexity is attainable by fusion of sensors and machine learning techniques. Identifying the right combination of sensory information and the appropriate machine learning technique is a vital ingredient for translating physical actions to virtual movements. The contributions of our work include: (i) Synchronization of actions and movements using suitable multiple sensor units, and (ii) selection of the significant features and an appropriate algorithm to process them. This work proposes an innovative approach that allows users to move in virtual environments by simply moving their legs towards the desired direction. The necessary hardware includes only a smartphone that is strapped to the subjects’ lower leg. Data from the gyroscope, accelerometer and campus sensors of the mobile device are transmitted to a PC where the movement is accurately identified using a combination of machine learning techniques. Once the desired movement is identified, the movement of the virtual avatar in the virtual environment is realized. After pre-processing the sensor data using the box plot outliers approach, it is observed that Artificial Neural Networks provided the highest movement identification accuracy of 84.2% on the training dataset and 84.1% on testing dataset.

In the field of Cyber Threat Intelligence (CTI), the scarcity of high-quality and labelled datasets that include Indicators of Compromise (IoCs) impact the design and implementation of robust predictive models that are capable of classifying IoCs in online communication, specifically in social media contexts where users are potentially highly exposed to cyber threats. Thus, the generation of high-quality synthetic datasets can be utilized to fill this gap and develop effective CTI systems. Therefore, this study aims to fine-tune OpenAI’s Large Language Model (LLM), Gpt-3.5, to generate a synthetic dataset that replicates the style of a real social media curated dataset, as well as incorporates select IoCs as domain knowledge. Four machine-learning (ML) and deep-learning (DL) models were evaluated on two generated datasets (one with 4000 instances and the other with 12,000). The results indicated that, on the 4000-instance dataset, the Dense Neural Network (DenseNN) outputs the highest accuracy (77%), while on the 12,000-instance dataset, Logistic Regression (LR) achieved the highest accuracy of 82%. This study highlights the potential of integrating fine-tuned LLMs with domain-specific knowledge to create high-quality synthetic data. The main contribution of this research is in the adoption of fine-tuning of an LLM, Gpt-3.5, using real social media datasets and curated IoC domain knowledge, which is expected to improve the process of synthetic dataset generation and later IoC extraction and classification, offering a realistic and novel resource for cybersecurity applications.

Predicting student performance in Virtual Learning Environments (VLEs) has become increasingly important with the growth of online education. Early identification of at-risk students enables timely interventions to improve academic outcomes. This study evaluates the performance of several Deep Learning (DL) models for tabular data, including ResNet, NODE, AutoInt, TabNet, TabTransformer (TT), SAINT, and GatedTabTransformer (GTT). Moreover, it examines the role of resampling techniques, including SMOTE, ROS, ADASYN, RUS, and Tomek-Links, in addressing class imbalance. Using the OULA dataset, eight experiments were conducted for binary and multi-class classification tasks, testing different feature combinations: (1) behavioral; (2) demographic and behavioral; (3) academic and behavioral; and (4) demographic, academic, and behavioral. The results indicate that the incorporation of a comprehensive set of characteristics significantly enhances the performance of the model, with academic characteristics proving more predictive than demographic characteristics. The SAINT model achieved the highest performance in binary classification (94.33% accuracy), leveraging its ability to capture simple yet meaningful feature interactions. For multi-class classification, SAINT again outperformed other models, achieving an accuracy of 73.22% when using Tomek-Links method, excelling in managing complex feature interactions and underrepresented classes such as “Distinction”. Statistical analysis using the Friedman aligned ranks test and the Nemenyi post-hoc test was conducted to compare model performance based on F1-scores from multiple experimental runs. The non-parametric Friedman test revealed significant differences among the models (p = 0.00013). SAINT and AutoInt consistently outperformed the other approaches, while ResNet and TT demonstrated the weakest performance. Post-hoc analysis using the Nemenyi test showed no statistically significant differences among mid-tier models (TabNet, GTT, NODE). A critical difference (CD) further confirmed SAINT and AutoInt as the most effective architectures for addressing complex, imbalanced educational data. These findings highlight the importance of aligning model selection and resampling techniques with the complexity of the task and the characteristics of the data.