Malware detection is a significant factor in establishing effective cybersecurity in the face of constantly increasing cyber threats. This research article aims to investigate the field of machine learning (ML) techniques for malware detection. More specifically, the paper focuses on the Customized K-Nearest Neighbors (C-KNN) classifier and the Firefly Algorithm (FA). The work aims to assess the effectiveness of C-KNN and C-KNN with FA (C-KNN/FA) in malware identification using the MalMem-2022 dataset. The novelty of the proposed method lies in the synergistic integration of the C-KNN algorithm with the FA for metaheuristic optimization. The use of FA to select the most relevant features enables the C-KNN to train on a small and high-quality feature set. Therefore, the performance of malware detection will be improved. We compare the performance of both methods to understand the influence of KNN parameter adjustment and feature selection on malware classification. The C-KNN and C-KNN/FA have produced remarkable results in malware identification, reaching an accuracy of 99.98%. This accomplishment is quite encouraging. With regard to multiclass and binary classification methods, C-KNN and C-KNN/FA both perform better than their alternatives.

This study investigates the performance of students enrolled in programming courses offered under the Sandwich Educational System (SES) in Ghanaian universities and colleges. SES is a unique educational approach that combines academic studies with practical work experience. This study examines various teaching models employed within the SES for programming education to identify any significant relationship between teaching methods and student academic performance. The target population for this study comprised students enrolled in computing education-related programs within the SES, with a specific focus on those undertaking programming courses. A single study group of students pursuing the Bachelor of Education programme in Information Technology (B.Ed. IT) under the sandwich mode at University X was selected to ensure efficient research management in this study. Employing a mixed-methods research design, quantitative and qualitative data were collected and analysed using descriptive and inferential statistics. A survey was administered to 218 of the 357 students in the study group during the designated survey period. Additionally, a seven-year longitudinal quasi-experiment involving five different year groups in the B.Ed IT sandwich programme at University X was conducted to examine the relationship between student performance and teaching methods within SES. The findings of this study do not demonstrate a significant difference in academic performance among students taught using different teaching methods in SES. However, it is crucial to acknowledge the study's limitations, which necessitate considering the findings as insightful observations rather than as conclusive results. This study recommends enhancing students' prior exposure to programming and adopting innovative teaching methods to improve their academic performance. Future research should address the limitations of this study by utilising a more rigorous experimental design, such as a randomised controlled trial, and exploring additional factors that may influence student performance within the SES. Such endeavours would enable more robust causal inferences to be drawn.

Deep-sea debris poses a significant threat to marine life and human health. Traditional methods for underwater debris detection and classification are labour-intensive and inefficient. The major challenge for using vision robots or autonomous underwater vehicles(AUVs) to remove deep sea debris is to exactly identify the marine debris. Marine debris gets deformed, eroded, and blocked due to seawater. Marine debris changes its shape, size, and texture in sea environment. Sea environment is challenging for the task of debris detection because of weak light. Uncertainty about the task of debris detection is due to marine life, rocks, marine flora, fauna, algae, etc. This study aims to develop a robust deep learning model for underwater debris detection and classification using YOLOV8. We evaluate the performance of YOLOV8 against YOLOV3 and YOLOV5 on the JAMSTEC TrashCan dataset. By employing an anchor-free detection head, YOLOV8 demonstrates improved accuracy in detecting underwater debris of varying shapes, sizes, and textures. Here, we show that YOLOV8 achieves a mean Average Precision (mAP) of 0.5095, outperforming YOLOV3 (mAP: 0.31879) and YOLOV5 (mAP: 0.43608). Our findings underscore the potential of anchor-free YOLOV8 in addressing the challenges of underwater debris detection, which is crucial for marine conservation efforts.

A generalized deep learning approach tracking image forgeries of any category with reduced architectural complexity, without compromising the performance is presented in this paper. A convolutional encoder-decoder architecture-based image reconstruction model is framed to extract all the pertinent information from the images. Performance comparison of similar networks constructed with varying architectural complexity led to the selection of this design. The best reconstruction feature extractor showed faster convergence and improved accuracy, as observed from the training and validation performance curves. Dimensionally compressed information from the reconstruction model is utilized by dense layers and further classified. Experimenting with forgery datasets inclusive of different forgery types ensured the generalizability of the model. In comparison with the reconstruction models adopting transfer learning in the encoder side utilizing MobileNet, ResNet 50, and VGG 19, the proposed model exhibited competitive and consistently improved mean Precision and F1-score performance across multiple datasets, as validated through multi-seed experimentation. Additionally, with the reduced architecture, the proposed model performed on par than the state-of-the-art approaches against which it was compared.

This research discusses consumer behavior toward unplanned purchasing, or what is popularly referred to as impulsive buying, in an online setting. The variables considered are Fear of Missing Out (FoMO), Price Perception, and Hedonism, and their effect on purchasing habits, with live streaming as a mediating variable for Generation Z TikTok Shop users. FoMO is a psychological state of anxiety regarding missing out on experiences that are appreciated by others. Price Perception refers to the act of deciphering price options based on obtained information in order to develop a sense of a product's pricing. Hedonism, on the other hand, refers to the tendency to seek pleasure and enjoyment. Generation Z, with its technological and social media savviness, often exhibits spontaneous buying tendencies in the case of online shopping, particularly on highly used apps like TikTok Shop. This research is quantitative in nature with survey techniques on 243 respondents of Generation Z who are active on TikTok Shop in Jakarta, Bogor, Depok, Tangerang, and Bekasi—locations that significantly contribute to the population and economic growth of Indonesia. The findings show that there is a positive correlation between FoMO and Price Perception variables and online impulsive buying, while Hedonism did not have any effect. In addition, Generation Z's engagement with live streaming shows a substantial positive effect on all three independent variables associated with impulsive consumer behavior. The results offer essential insight for e-commerce professionals who aim to create more effective marketing strategies by utilizing drivers that can strengthen and stimulate impulsive buying behavior among Generation Z consumers.

Expansion of Internet of Things (IoT) technologies has greatly enhanced monitoring and management of energy systems, especially in Hybrid Renewable Energy Systems (HRES). This paper presents an IoT-based HRES smart grid framework with a modified Brain Storm Optimization (BSO) algorithm for routing optimization and an Improved Quantum Key Management (IQKM) is a quantum inspired protocol for better data security. The enhanced BSO algorithm, hosted in the cloud infrastructure, optimizes IoT sensor data routing paths, thus diminishing packet transmission latency and improving the network throughput. In contrast to conventional BSO techniques, the enhancement is through dynamic cluster refinement and adaptive node prioritization, designed specifically for real-time cloud-integrated energy systems. In order to protect sensitive energy transmission information, the IQKM protocol includes strong quantum-aided encryption processes and dynamic key creation. These enhancements directly counter the dangers of man-in-the-middle and replay attacks, which exceed capabilities of standard encryption approaches by facilitating low-latency, quantum-resistant communication between HRES nodes. Both Photovoltaic (PV) and wind-based energy sources are utilized by the system to provide power consistently, with cloud-based analytics and IoT sensors ensuring real-time monitoring. Experimental testing via the Adafruit platform reports a 23% Packet Delivery Ratio (PDR) enhancement and 17% encryption/decryption delay reduction compared to baseline and traditional routing algorithms. Such findings ensure the potential for stable, secure, and scalable grid performance by the proposed system.

The rapid growth of IoT ecosystems has intensified the complexity of fog–cloud infrastructures, necessitating adaptive and energy-efficient task offloading strategies. This paper proposes MADRL-MAML, a Multi-Agent Deep Reinforcement Learning framework enhanced with Model-Agnostic Meta-Learning for dynamic fog–cloud resource allocation. The approach integrates curriculum learning, centralized attention-based critics, and KL-divergence regularization to ensure stable convergence and rapid adaptation to unseen workloads. A unified cost-based reward formulation is used, where less negative values indicate better joint optimization of energy, latency, and utilization. MADRL-MAML is benchmarked against six baselines Greedy, Random, Round-Robin, PPO, Federated PPO, and Meta-RL using consistent energy, latency, utilization, and reward metrics. Across these baselines, performance remains similar: energy (3.64–3.71 J), latency (85.4–86.7 ms), and utilization (0.51–0.54). MADRL-MAML achieves substantially better results with a reward of $-21.92 \pm 3.88$, energy 1.16 J, latency 12.80 ms, and utilization 0.39, corresponding to 68\% lower energy and 85\% lower latency than Round-Robin. For unseen workloads characterized by new task sizes, arrival rates, and node heterogeneity, the meta-learned variant (MADRL-MAML-Unseen) achieves a reward of $-6.50 \pm 3.98$, energy 1.14 J, latency 12.76 ms, and utilization 0.73, demonstrating strong zero-shot generalization. Experiments were conducted in a realistic simulated environment with 10 fog and 2 cloud nodes, heterogeneous compute capacities, and Poisson task arrivals. Inference latency remains below 5 ms, confirming real-time applicability. Overall, MADRL-MAML provides a scalable and adaptive solution for energy-efficient and latency-aware orchestration in fog–cloud systems.

This paper explores the use of stochastic optimization techniques to address the aircraft allocation problem under uncertain passenger demand. The proposed stochastic allocation model successfully meets the study’s objectives by demonstrating how uncertainty in passenger demand can be effectively incorporated into aircraft assignment decisions through a two-stage stochastic programming framework. Simulation results across multiple demand scenarios show that the model provides stable and adaptive allocations that minimize total cost while maintaining service quality, even under high variability. Incorporating the simple recourse approach enables post-decision flexibility, reducing penalties for unmet demand, and the use of Geometric Brownian Motion (GBM) offers a realistic representation of continuous demand fluctuations over time. These outcomes confirm the model’s practical value in bridging deterministic planning and real-time decision environments. While future research will focus on extending the model to a Markov Decision Process (MDP) framework and integrating real-time data streams, the current results establish a solid foundation by quantifying how uncertainty directly impacts fleet utilization, cost efficiency, and service reliability.

Early prediction of students' placement outcomes is critical for aligning curricula with industry demands, optimizing academic planning, and providing focused career support. It also enhances institutional reputation, strengthens employer partnerships, and supports data-driven decision-making. However, predictive modeling in this context is challenged by data heterogeneity, evolving market factors, subjective evaluations, and bias mitigation. This study proposes an AI-driven framework that integrates Gated Recurrent Unit (GRU) networks with Modified Dwarf Mongoose Optimization (MDMO) to address these challenges. GRU effectively captures temporal patterns in academic and behavioral data, while MDMO ensures optimal hyperparameter tuning through advanced search strategies. Model performance was rigorously evaluated using multiple metrics including accuracy, false positive rate (FPR), false negative rate (FNR), sensitivity, specificity, and Matthews Correlation Coefficient (MCC). The proposed GRU-MDMO model achieved an accuracy of 98.5%, sensitivity of 97.78%, specificity of 99.09%, and MCC of 96.97%, outperforming other baseline models such as SVM, ANN, RF, and traditional GRU variants. These results demonstrate the model’s robustness, reliability, and suitability for early placement prediction. This approach empowers institutions to improve placement rates, enhance curriculum design, attract admissions, and ultimately foster better student career outcomes through AI-guided educational intelligence.

Cloud computing can be revolutionized by quantum computing which will offer the world more computational power than has ever been seen to solve complex issues. Quantum computing coupled with cloud computing enables the remote access to quantum resources, thus greatly minimizing the cost, technical, and operational difficulties of having quantum hardware owned and maintained in the field. The integration makes large-scale data processing, cryptography, and optimization tasks as well as new applications in artificial intelligence efficient in terms of their computation. This work is a review of the existing approaches, system, and systems to quantum cloud computing, the main algorithms, software applications, implementation plans, and real-life examples. We find that quantum cloud computing provides significant enhancements in computational speed and parallelism, scalability, as well as provides the capability to process data securely and to execute quantum circuits remotely. However, there are still a few obstacles such as stability of qubits, error correction, noise reduction, and effective resource utilization, which restrict the practical use of quantum cloud services. The findings indicate that, irrespective of these challenges, quantum computing with the use of cloud computing platforms offers meaningful potentials to scientific discovery, business, and an AI-based innovation. The paper wraps up by noting that further research should be done to enhance the reliability of quantum hardware, optimize quantum algorithms, and design quantum cloud computing security systems, enabling quantum cloud computing to be adopted more broadly as a more transformative model of computation and ensuring that quantum cloud computing can grow sustainably.