As the Multimedia Internet of Things (M-IoT) evolves, the orchestration of numerous resources that offer support for high-bandwidth, low-latency applications arises as a key challenge. Architecturally, the edge-cloud framework alleviates structural concerns, but the linked nature of compute and data transfer poses problems of resource management. Approaches that tackle task offloading and adaptive transmission that think independently of each other tend to have problems such as user-server cross-region overloads or network congestion. This paper presents JATO, a framework to jointly tackle the problems of adaptive task offloading and transmission optimization using Deep Reinforcement Learning. JATO offers a mono-faceted solution, learning a policy to simultaneously determine the best offloading target and the transmission quality. The framework was implemented for evaluation with a combination of different edge devices in a testbed alongside a simulation environment. JATO recorded a result of 0.9321 as the holistic score of the overall framework endpoint, a score significantly better than that of all the other frameworks that were used as functional baselines. JATO was able to resource optimally with a network lag of 131.65 milliseconds and a network freeze of 0.09% with the resources utilized. This is evidence that offloading and rate control in combination provides better resource elasticity for M-IoT systems.
In this work, a comparative numerical simulation of single-gate and double-gate depletion mode GaAs MESFETs in the nanometer domain using COMSOL Multiphysics was carried out. The objective was to evaluate the significance of multi-gate structures based on electrostatic control capability in terms of performance improvement in Schottky gate MESFETs. Four models were considered, including a traditional single-gate model, an analogous double-gate model, and their corresponding reduced scale models. Comparisons were made in terms of the concentration of electrons and output response depending on the value of the gate bias voltage. There were almost no differences between the two models when scaled at the same dimensions. The reduced dual-gate device exhibited lower leakage current and superior channel control compared to its single-gate counterpart, particularly with higher drain bias applied. This suggests that while the dual-gate design may not be advantageous in all dimensions, it proves more effective when MESFET structures are aggressively scaled, indicating that dual-gate configurations are preferable for nanoscale GaAs MESFETs operating in regimes where electrostatic degradation is significant.
The integration of terrestrial 5G/6G networks with Low Earth Orbit (LEO) satellite systems is creating a class of hybrid Satellite–Terrestrial Integrated Networks/Non-Terrestrial Networks (STIN/NTN) for which channel management in the face of highly dynamic topologies and traffic heterogeneity is becoming a key challenge. This paper examines intelligent approaches to radio resource allocation in such networks based on deep reinforcement learning (DRL) and multi-agent algorithms. A conceptual architecture for a channel management system in a hybrid network with a LEO segment is proposed, including a centralized Software-Defined Networking/Artificial Intelligence (SDN/AI) coordinator and distributed DRL agents at the satellite and ground base station levels. To analytically interpret the gains from intelligent management, an aggregated M/M/c/c teletraffic model is introduced, using the concept of an “equivalent number of channels,” which allows for linking DRL resource allocation with classical assessments of blocking probability and throughput. The obtained numerical results show that increasing the effective system capacity by 20% due to intelligent inter-segment load redistribution leads to a 30–40% reduction in blocking and a 10–18% increase in throughput in the high-load region, which is consistent with the results of detailed DRL studies for LEO networks presented in the modern literature.
JavaScript is highly demanded in modern development, but traditional learning methods often fail to engage students. To address this, this study designed and developed a responsive, gamified web-based JavaScript learning application using the ReactJS library. The application integrates gamification elements such as dashboards, learning modules, challenges, points, levels, ranks, and achievements to enhance learning engagement. The application’s usability and acceptance were evaluated using the Technology Acceptance Model (TAM) through a survey involving 35 participants. The evaluation yielded a Perceived Usefulness score of 90.00% and a Perceived Ease of Use score of 89.71%, resulting in a high overall average score of 89.85%. These results indicate that respondents strongly agree that the application is highly useful for escalating learning motivation and is easy to use. While the TAM results confirm high user acceptance and demonstrate that gamification effectively eases the JavaScript learning process, our future research will incorporate objective cognitive assessments to definitively measure its absolute pedagogical effectiveness.
Traditional communication systems are often disrupted during disasters, leaving communities disconnected and interrupting the exchange of essential information. In these circumstances, a reliable communication approach is critical for coordinating relief and recovery activities. This work presents a Resilient Communication Model (RCM), an infrastructure-free solution designed to enhance post-disaster communication. The model relies on dynamic, self-organizing links between nearby devices, allowing victims and rescuers to communicate effectively. Due to its lightweight and flexible design, the system is particularly well-suited for basic devices, which are more likely to remain functional in crisis conditions. Simulation studies and comparative evaluations with other algorithms demonstrate that the model minimizes packet loss ratios (PLR), delay, and enhances communication stability, packet delivery ratio (PDR) and throughput during network disruptions. Practical experiments across diverse scenarios further confirm the model’s feasibility and resilience. Overall, the findings demonstrate that the model can reliably function as a fallback communication system when standard networks are unavailable.
Fast, responsive, and informative public services are societal demands that must be fulfilled by government agencies, among which the Department of Population and Civil Registration of Surabaya City. To enhance service quality, this study developed a Large Language Model (LLM)-based Question Answering (QA) system to address public inquiries regarding Identity Card (ID) and Family Card (FC) services. The proposed system utilizes the Komodo-7B model, which was customized using Quantized Low-Rank Adaptation (Q-LoRA) fine-tuning and integrated with a Retrieval-Augmented Generation (RAG) approach to improve the accuracy and relevance of the generated responses. The training process leveraged a real-world complaint dataset from Disdukcapil alongside the open-source MS MARCO dataset. Furthermore, the RAG implementation employs sentence vectorization via SentenceTransformer and cosine similarity-based context retrieval. System performance was evaluated using ROUGE and METEOR metrics across four scenarios: Komodo-7B Base, RAG Komodo-7B Base, Fine-Tuned Komodo-7B, and RAG Fine-Tuned Komodo-7B. The results show that the RAG Fine-Tuned Komodo-7B configuration delivered the best performance, achieving F1-Scores of 0.3554 for ROUGE-1, 0.3096 for ROUGE-L, and 0.2886 for METEOR.
Intrusion detection in IoT-enabled cloud environments is challenged by high-dimensional traffic, class imbalance, and limited labeled data. This paper proposes a hybrid framework combining Golden Jackal–Grey Wolf Optimization (GJO-GWO) for feature selection with a Kernel Mean Alignment Autoencoder (KMA-AE) for deep transfer learning. GJO-GWO selects a compact, discriminative feature subset, while KMA-AE aligns source and target latent representations to mitigate distribution mismatch. Experiments on the CIDDS-001 dataset achieve 90.21% accuracy and 0.90 macro-F1, with improved precision–recall for minority attacks and a 60% feature reduction. Although training is more expensive, the method attains the lowest inference time, enabling real-time deployment. Overall, the framework provides an effective and generalizable intrusion detection solution for dynamic IoT environments.
E-commerce platforms are increasingly challenged to sustain customer engagement amidst intensifying competition. Traditional gamification approaches, characterized by static, uniform mechanics, often fail to adapt to individual user preferences, leading to diminishing returns and decreased engagement over time. These conventional methods typically employ fixed reward structures that do not account for individual user behavior, resulting in a lack of sustained engagement. This paper introduces a comprehensive machine learning (ML) framework for dynamic gamification, designed to personalize game elements in real-time based on individual user behavior patterns. The framework integrates clustering algorithms, reinforcement learning (RL), and collaborative filtering techniques to analyze user interactions and generate adaptive gamified experiences. Simulated testing, conducted using a publicly available e-commerce customer behavior dataset from Kaggle, provided insights into diverse user preferences and behaviors. Simulated results demonstrated significant improvements, including a 32% increase in daily active users, a 24% higher conversion rate, and a 30.8% improvement in 30-day customer retention. The framework addresses critical technical challenges, such as scalability, real-time processing, and ethical data usage. This research contributes to the advancement of personalized digital experiences in e-commerce, offering practical guidelines for enhancing customer engagement through AI-driven gamification.
Mobile health applications (mHealth) are known for their potential to enhance health delivery and improve health outcomes for older adults. However, despite their broad potential, significant usability concerns persist among older adults, preventing them from gaining advantages from mHealth and hindering their adoption. This paper critically examines the key usability concerns older adults face when interacting with mHealth. This research adhered to the PRISMA guidelines in the process of retrieving and selecting articles, resulting in the review and analysis of 117 articles published between 2014 and 2024. The results highlight several prevalent usability concerns, including complex navigation, overwhelming instructions, inadequate feedback mechanisms, and limited personalization options, which overlook the diverse needs of older users. Furthermore, this study identified design solutions to address these concerns, including straightforward navigation, clear instructions, and consistent layout/design tailored to older adults. Moreover, this paper established a visual representation that maps identified usability challenges to corresponding design solutions, serving as a valuable resource for researchers, designers, and developers working on mobile health applications for older adults. Integrating these solutions into the development process can significantly improve user experience and increase engagement in health interventions.
This study proposes a novel application of Particle Swarm Optimization (PSO) for tool path planning in complex louvre geometries utilized in heat transfer systems. Unlike conventional approaches, the proposed framework explicitly integrates geometric smoothness and collision avoidance into the optimization process, as it enables the generation of continuous and non-intersecting tool trajectories. This is particularly significant as surface quality in louvre fins directly influences boundary layer disruption, which in turn affects convective heat transfer efficiency and pressure drop characteristics. By minimizing abrupt tool movements and machining-induced surface roughness, the method addresses a critical gap between manufacturing precision and thermal performance. The PSO-based approach simultaneously optimizes machining time and trajectory feasibility, ensuring safe and efficient tool movements. The experimental results demonstrated rapid convergence with the objective function significantly decreasing within the first 50 iterations and stabilizing around iteration 80. The optimized solution achieved a machining time of 0.60 minutes (36 seconds) while maintaining consistent minimum objective values throughout the process. These findings highlight the robustness and stability of the proposed method. Overall, this work contributes a novel optimization framework that bridges advanced manufacturing and thermal performance considerations, establishing PSO as an effective solution for high-precision tool path planning in complex industrial geometries.
This paper discusses asset challenges detection in the healthcare industry, specifically delays and inaccuracies in asset monitoring caused by suboptimal RFID polling methods. The research question is how to determine an appropriate RFID polling interval that balances asset location accuracy, reader energy consumption, and network response time. Due to the increasing risk of mismanagement and equipment loss, an efficient approach is needed to improve asset-tracking accuracy. This study proposes a simulation-based multi-objective optimization approach by determining the optimal polling period to minimize network delay, reader energy consumption, and false identifications. Monte Carlo simulation models the stochastic movement of assets to evaluate system performance under different polling strategies. The results of one experiment showed that 100 assets, with an average moving rate of 2.48, reached the optimal scanning period of 1460 minutes. Additional experiments were conducted to analyze the sensitivity of the optimal polling interval to changes in asset population and movement rates. The contribution of this study is the development of a holistic model to determine the optimal scanning time to improve asset-tracking accuracy and reduce operational costs in RFID systems. Although evaluated in a healthcare context, the proposed framework is versatile and can also be used for other RFID-based asset monitoring scenarios with similar trade-offs.
The growing volume of global waste poses significant challenges to effective waste management, underscoring the need for innovative classification methods to improve recycling efficiency. This study evaluates the performance of two traditional machine learning models, Naïve Bayes and Support Vector Machines (SVMs), for classifying solid waste materials in an automated sorting system. A dataset of 284 JPEG images, categorized into five classes (cardboard, glass, metal, paper, and plastic), was utilized. Preprocessing involved resizing images to 512x384 pixels, normalizing pixel values, and extracting features using Histograms of Oriented Gradients (HOG) and Color Histograms. Naïve Bayes demonstrated computational efficiency with 98.90% accuracy and an F1-score of 0.908, but struggled with overlapping features, leading to misclassifications, particularly between glass and metal. In contrast, SVM outperformed Naïve Bayes, achieving 99.80% accuracy and an F1-score of 0.965 by effectively handling complex, overlapping features via optimal decision boundaries. The findings highlight SVM’s superior performance for complex datasets, while Naïve Bayes remains a viable option for simpler tasks. This study underscores the potential of traditional machine learning in waste classification. However, it suggests that integrating deep learning models could improve accuracy, scalability, and adaptability in real-world waste-sorting systems.
The increasing demand for distilled water (Aquades) in pharmaceutical and medical applications contrasts sharply with the limited quality of municipal water supplies and the high operating costs of commercial Aquades procurement. At the same time, many small-scale facilities still lack integrated systems capable of meeting the Indonesian Ministry of Health standard (Permenkes RI No. 32/2017). Existing research on reverse osmosis (RO) systems largely focuses on membrane or filtration performance, with limited attention to real-time water-quality monitoring and systematic optimization of pre-treatment filters. This study develops an integrated filtration and monitoring system designed to ensure regulatory compliance while optimizing the composition of pre-treatment materials. The system combines silica sand, activated carbon, and zeolite pre-filters with RO, supported by six analog sensors that continuously monitor pH, turbidity, and Total Dissolved Solids before and after filtration. Validation results show high sensor accuracy, with 99.77% for TDS, 98.10% for pH, and 99.97% for turbidity. Among six tested filter compositions, the 25% silica sand-25% activated carbon-50% zeolite configuration achieves the highest average filtration efficiency of 88.96%. These findings demonstrate that optimized pre-treatment combined with real-time monitoring can significantly improve RO performance and support cost-effective Aquades production for medical use.
This study proposes a scalable and efficient approach for predicting student behaviour in large-scale educational environments. It introduces a parallelized hybrid model that combines Density-Based Optimized K-Means clustering, Analytic Hierarchy Process (AHP) feature weighting, and Hierarchical K-Nearest Neighbours (KNN), implemented using Apache Spark. The main research question is how to improve scalability, accuracy, and computational efficiency of student behaviour prediction when dealing with large, complex datasets. The model addresses key limitations of traditional methods, such as handling heterogeneous data, treating all features equally, and high computational cost. Two main innovations are presented. First, AHP is used to assign structured importance to features, allowing critical factors like attendance and study time to have greater influence on prediction accuracy. Second, clustering and prediction are parallelized using Spark, enabling efficient real-time processing of large datasets. The approach was evaluated using 18,586 student records and more than 20 million behavioural entries. Results show that Hierarchical KNN consistently outperforms standard KNN as dataset size increases. While traditional KNN shows unstable error rates, peaking at 9.4%, Hierarchical KNN maintains lower and more stable errors between 5.16% and 6.08%. Execution time was also significantly reduced through parallel processing, though gains were limited by communication overhead. Overall, the proposed model offers a robust framework for real-time behaviour analysis, academic risk detection, and targeted educational intervention.
Breast cancer (BC) is becoming a global epidemic, largely affecting women. Breast cancer cases keep climbing steadily. Thus, early detection technologies or systems that notify patients to this disease are essential. Individuals can start treatment for this life-threatening illness, so that patients may be cured or given longer lives. To achieve this, in this study, an expert intelligence system named Intelligent System for Predicting Breast Cancer (ISPBC) was developed. The proposed system utilizes an innovative feature selection technique known as Enriched Feature Set (EFS) in order to identify the most appropriate and significant features. The proposed EFS employs the advantages of heuristic search techniques and stochastic hill climbing to select the most significant and important features. The Decision Tree and Random Forest techniques are employed for breast cancer diagnosis, distinguishing between malignant and benign types. The suggested model’s performance was evaluated by comparing measures such as accuracy, precision, and recall through the utilization of tenfold cross-validation. To measure the efficacy of the suggested model, ISPBC’s performance was compared to that of base classifiers and models published in the literature. A maximum accuracy of 96.09% was attained by ISPBC according to the results.
This study presents a new approach for constructing a medical knowledge graph using Named Entity Recognition (NER) to identify entities such as diseases, drugs, or medical procedures, alongside part-of-speech (POS) tagging and dependency parsing to determine words that function as verbs and roots. These extracted words are then used as relations between entities, forming triplets in the format (entity, relation, entity). While the knowledge graph provides a structured representation of medical information, the evaluation primarily reflects the performance of the underlying NLP pipeline (NER, POS tagging, and dependency parsing) used to generate the triplets. Quantitative evaluation was performed using metrics such as precision, recall, and F1-score to assess the accuracy and completeness of entity and relation extraction. The qualitative evaluation involved medical domain experts to assess the relevance and validity of the relationships derived. The results indicate that fine-tuning a pre-trained model for NER and leveraging a pre-trained model for POS tagging and dependency parsing can effectively generate accurate triplets for constructing a medical knowledge graph. This approach demonstrated strong performance, achieving high evaluation scores in both quantitative and qualitative evaluations.
The expansion of communication technology and the increasing usage of the frequency spectrum drive the need for compatible device testing. Wideband antennas play a crucial role in supporting modern communication systems and applications, including those used as the sensors in electromagnetic compatibility (EMC) testing. Optimization techniques, such as genetic algorithm (GA), particle swarm optimization (PSO), and covariance matrix adaptation–evolution strategy (CMA-ES), are widely applied to enhance the bandwidth of electromagnetic devices. However, most studies focus on individual algorithms or limited comparisons, resulting in a lack of systematic evaluation within a unified framework. This paper fills that gap by directly comparing GA, PSO, and CMA-ES on the same planar sensor design, assessing their effectiveness in achieving the widest bandwidth. The planar sensor had a basic spline-based configuration using quadratic Bezier equation. A performance comparison based on a simulation showed that the planar sensor configuration with the best bandwidth was 17.77 GHz, spanning a frequency range from 2.23 GHz to 20 GHz, which was limited by the highest observation frequency of the available measuring instrument. Furthermore, verification of the realized planar sensor showed that the bandwidth reached 17.86 GHz, from 2.14 GHz to 20 GHz, with a geometric bandwidth of 273%.
Large language models (LLMs) have undergone rapid evolution and are highly effective in tasks such as text generation, question answering, and context-driven analysis. However, the unique requirements of Islamic studies, where textual authenticity, diverse jurisprudential interpretations, and deep semantic nuances are critical, present challenges for general LLMs. This article reviews the evolution of neural language models by comparing the historical progression of general LLMs with emerging Islamic-specific LLMs. We discuss the technical foundations of modern Transformer architectures and examine how recent advancements, such as GPT-4, DeepSeek, and Mistral, have expanded LLM capabilities. The paper also highlights the limitations of standard evaluation metrics like perplexity and BLEU in capturing doctrinal, ethical, and interpretative accuracy. To address these gaps, we propose specialized evaluation metrics to assess doctrinal correctness, internal consistency, and overall reliability. Finally, we outline a research roadmap aimed at developing robust, ethically aligned, and jurisprudentially precise Islamic LLMs.
The fifth-generation (5G) mobile communication system represents a major advancement in wireless technology, relying on effective radio spectrum management to ensure optimal performance. Among the available frequency ranges, the 5G low-band spectrum provides extensive coverage but limited capacity, making its efficient management a critical challenge. This study presents a predictive model based on the system dynamics approach to analyze the management of the 5G low-band spectrum. The model captures the interrelationships between technical and economic variables that influence spectrum allocation and service adoption over time. Three simulation scenarios—low, medium, and high allocation rates—were developed to examine allocation patterns and their effects on 5G service diffusion. The results revealed that spectrum management in 5G exhibits goal-seeking behavior constrained by spectrum scarcity, with service adoption showing a growth-to-saturation pattern. The findings demonstrate that appropriate low-band spectrum management can significantly enhance 5G deployment efficiency. The proposed model serves as a decision-support tool for policymakers and regulators, enabling evaluation of alternative management strategies prior to policy implementation and promoting evidence-based decision-making in future 5G spectrum policies.
The pace and complexity of modern cyber-attacks expose the limits of traditional ‘impact × likelihood’ risk matrices, which compress uncertainty into coarse categories and miss inter-dependent threat dynamics. We propose a three-layer multi-fuzzy inference system (MFIS) that models general infrastructure vulnerabilities and access-control weaknesses separately, then fuses them into a single, continuous 0-25 risk score. The framework was validated on three representative scenarios—catastrophic/continuous, serious/frequent, and minor/few attacks—encompassing sixteen threat criteria. Compared with a crisp 5 × 5 matrix, MFIS cut mean-absolute error and root-mean-square error by 90 to 99% and reproduced expert-panel judgments to within 0.55 points across all scenarios. Nine independent practitioners rated the prototype highly on usability (100% agreement), credibility (100%) and actionability (100%), with 78% willing to recommend adoption. These results demonstrate that MFIS delivers fine-grained, expert-aligned assessments without adding operational complexity, making it a viable drop-in replacement for time- or resource-constrained organizations. By capturing partial memberships and cross-domain interactions, MFIS offers a more faithful, adaptive and explainable basis for prioritizing cyber-defense investments and can be extended to emerging threat domains with modest rule-base updates.
