Garuda - Garba Rujukan Digital

Edge Computing Frameworks for Real-Time Optimisation in Autonomous Electric Vehicle Networks Ismail, Laith S.; Jamil, Abeer Salim; Ali, Taghreed Alaa Mohammed; Al-Dosari, Ibraheem Hatem Mohammed; Salman, Khdier; Maidin, Siti Sarah
International Journal of Engineering, Science and Information Technology Vol 5, No 3 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i3.1397

Autonomous electric vehicles (AEVs) require real-time decision-making, low-latency computation, and energy-aware coordination to operate effectively. Traditional centralised cloud computing struggles to meet these demands due to inherent delays and scalability issues in large-scale AEV networks. This paper proposes a novel hybrid edge–fog computing architecture to address these challenges. Our framework utilises a three-tier system (vehicle-edge, roadside-fog, and cloud) governed by a deep reinforcement learning agent that manages energy-aware task offloading. Extensive simulations demonstrate the framework's effectiveness, achieving significant end-to-end latency reductions of up to 56% during urban peak hours and decreasing energy consumption by 20% under high-load conditions. The deep reinforcement learning agent successfully adapts control policies to dynamic road conditions, while the architecture proves highly scalable and resilient, maintaining high task success rates and recovering from node failures in seconds. These findings confirm that a hybrid edge–fog architecture, guided by reinforcement learning, is a highly effective solution for scalable, adaptive, and energy-efficient AEV operations. This study's primary contribution is an empirically validated framework that uniquely integrates predictive control and energy-aware scheduling at the edge, providing a deployable model for next-generation intelligent transportation systems.

A Multiple Linear Regression Approach to Predicting AI Professionals’ Salaries from Location and Skill Data Maidin, Siti Sarah; Yi, Ding; Ayyasy, Yahya
International Journal of Informatics and Information Systems Vol 7, No 3: September 2024
Publisher : International Journal of Informatics and Information Systems

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/ijiis.v7i3.213

The rapid growth of Artificial Intelligence (AI) industries worldwide has increased the demand for skilled professionals and highlighted the need to understand salary determinants in this sector. This study aims to analyze the factors influencing the compensation of AI professionals globally, with a particular focus on the effects of company location, experience level, and required technical skills. Using a dataset of 15,000 AI job postings collected from multiple countries, a Multiple Linear Regression (MLR) model was developed to identify predictive relationships between independent variables—location, experience, and skills—and the dependent variable, annual salary in U.S. dollars. Data preprocessing included one-hot encoding for categorical variables, standardization of numerical attributes, and vectorization of text-based skill descriptions. Model evaluation produced strong predictive results, with an R² of 0.82, a Mean Absolute Error (MAE) of 18,677 USD, and a Root Mean Squared Error (RMSE) of 25,704 USD. Statistical tests confirmed that company location and experience level significantly affected salary outcomes (p 0.05), while technical skills contributed only marginally. These findings suggest that structural factors such as geography and seniority play a more decisive role in determining AI salaries than specific technical competencies. The study concludes that MLR offers a transparent and interpretable analytical framework for exploring salary disparities in the global AI workforce. The results provide practical implications for organizations designing fair compensation policies, professionals assessing market value, and educators aligning training programs with evolving industry demands.

Algorithms and Modeling for Optimizing Sustainable Energy Systems Jaleel Maktoof, Mohammed Abdul; Shaker, Alhamza Abdulsatar; Nayef, Hamdi Abdullah; Taher, Nada Adnan; Yousif Al Hilfi, Thamer Kadum; Maidin, Siti Sarah
International Journal of Engineering, Science and Information Technology Vol 5, No 1 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i1.1457

The global transition toward sustainable energy necessitates intelligent, integrated solutions to overcome the intermittency of renewable sources. This paper presents and validates a comprehensive framework for optimising Hybrid Solar-Wind Energy (HSWE) systems by integrating advanced simulation, machine learning-based forecasting, and metaheuristic optimisation. Using meteorological and operational data from three distinct climate zones, we modelled and analysed a PV-wind-lithium-ion hybrid system. A neural network was employed for precise load forecasting, while Particle Swarm Optimisation (PSO) managed real-time resource allocation and storage dispatch. Comparative analysis reveals that the optimised hybrid system significantly outperforms standalone units, increasing energy production by up to 32%, improving overall energy efficiency to 92.3%, and reducing operational costs by over 36%. The simulation models demonstrated high fidelity, with predictions matching experimental field data with less than 1% error. Furthermore, the integration of predictive fault handling and intelligent load balancing enhanced system reliability, increasing the mean time between failures (MTBF) by over 70% and achieving 97.6% system availability. This research provides a validated, replicable framework for engineers and policymakers, demonstrating a practical pathway to developing efficient, economically viable, and resilient decentralised renewable energy infrastructure to meet global sustainability goals.

Analyzing the Evolution of AIGenerated Art Styles Using Time Series Analysis: A Trend Study on NFT Artworks Maidin, Siti Sarah; Yang, Qingxue; Samson, A Sunil
Journal of Digital Market and Digital Currency Vol. 2 No. 2 (2025): Regular Issue June 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jdmdc.v2i2.32

This study investigates the development of AI-generated art styles within the growing non-fungible token (NFT) market. Using time series analysis, the research identifies key trends and shifts in art styles from 2022 to 2024, revealing how various art forms, algorithms, and mediums evolved in response to technological advancements and market forces. Data was collected from a sample of 10,000 NFT artworks, categorized by creation date, style, and algorithm usage. Exploratory Data Analysis (EDA) techniques, including line graphs and heatmaps, were employed to visualize and interpret trends across different art styles and AI tools. Results indicate a significant increase in the popularity of styles like surrealism and realism, with deepdream and GANpaint algorithms being frequently associated with these styles. Stacked area charts further highlighted the proportional growth of art styles over time, providing insights into both short-term popularity spikes and long-term trends. The findings suggest that the integration of AI algorithms significantly influenced the rise of specific art genres, with certain algorithms correlating strongly with particular styles. Practical implications for artists and collectors include the potential for data-driven insights to guide creative choices and investment strategies. The study's limitations, such as the lack of broader market data, provide a foundation for future research to explore the intersection of AI-generated art, NFT marketplaces, and cultural influences. The paper concludes that AI and NFTs are reshaping the traditional art market, presenting new opportunities for creativity, ownership, and artistic value in a digital age.

Price Trend Prediction and Discount Optimization for Video Games in Online Stores Using XGBoost and Time-Series Analysis: A Data Mining Approach for Metaverse-Driven Market Insights Maidin, Siti Sarah; Yahya, Norzariyah
International Journal Research on Metaverse Vol. 2 No. 4 (2025): Regular Issue December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/ijrm.v2i4.41

This research explores the application of data mining techniques, specifically XGBoost, to predict game pricing trends and optimize discount strategies within the digital gaming market. Game prices are influenced by various factors, including production costs, market demand, and promotional strategies. This study analyzes historical pricing data from multiple online stores to identify key pricing patterns and factors that influence price changes over time. The model developed in this study predicts game prices by incorporating features such as retail price, discount percentages, past price trends (lags), and other time-based features. The findings reveal that retail price and recent price trends (e.g., 7-day rolling averages) are the most influential features in predicting future prices. Additionally, discount strategies significantly impact game sales, with certain discount ranges showing higher effectiveness in driving consumer purchases. The model also demonstrates variability in prediction accuracy, particularly at higher price points, highlighting the challenges of capturing complex price fluctuations in a dynamic digital marketplace. The significance of this study extends to the Metaverse market, where pricing and the use of digital assets like non-fungible tokens (NFTs) play a critical role. The model's application could aid in optimizing pricing strategies within virtual economies, enhancing both the consumer experience and retailer profitability. Future work includes integrating additional features such as user reviews and exploring its application to Metaverse game platforms. The practical implications of this research are significant for online game retailers looking to leverage data-driven insights for more effective pricing and promotional strategies.

AI-Driven Text Analysis and Generation for Green Energy Applications Ahmed, Saif Saad; Mahdi, Mohammed Fadhil; Hammad, Qudama Khamis; Mahdi, Ammar Falih; Alfalahi, Saad.T.Y.; Maidin, Siti Sarah
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.1745

The rapid growth of the green energy sector has produced a massive volume of textual data, creating significant challenges for information extraction and decision support. This study investigates the application of state-of-the-art Natural Language Processing (NLP) models, specifically BERT and GPT-4, to automate and enhance policy drafting, market analysis, and academic research clustering. We evaluated these models on a corpus of over 200,000 energy-related documents, using a structured computational workflow to measure performance on semantic coherence, factual reliability, and processing efficiency. The results demonstrate substantial improvements over manual methods. The AI-driven approach reduced policy drafting time by 39% and error rates by over 58%, while increasing semantic alignment to 93.5%. In market report synthesis, the models improved topic extraction accuracy by over 10% and reduced summary generation time by 38%. For academic literature, thematic clustering accuracy reached 92.3%, with a 44% reduction in processing time. These findings validate that fine-tuned NLP models can serve as powerful analytical tools in the sustainable energy domain, enabling institutions to navigate complex regulatory and technical information more effectively. By providing a practical demonstration of how automated NLP solutions can augment human expertise, this work contributes to the applied use of AI in achieving global green energy objectives, while also considering the associated methodological and ethical implications.

Blended Learning for Character Education: Integrating Tri Hita Karana Wisdom to Develop Graduate Competencies Aligned with SDGs Astawan, I Gede; Utama, I Dewa Gede Budi; Paramartha, Wayan Eka; Abdurahman, Ayi; Maidin, Siti Sarah
Journal of Educational Technology and Learning Creativity Vol. 3 No. 2 (2025): December
Publisher : Cahaya Ilmu Cendekia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37251/jetlc.v3i2.2498

Purpose of the study: This study examines the effectiveness of a technology-enhanced character education model based on the Balinese Tri Hita Karana philosophy in developing primary students’ competencies aligned with SDG 4. Using blended learning, it addresses gaps in research on culturally grounded technology integration in character education. Methodology: This quasi-experimental study involved 92 fifth-grade students from four primary schools in Karangasem, Bali, comparing a Technology-Enhanced Tri Hita Karana Character Education Model with conventional instruction. Using LMS, digital portfolios, gamification, and learning analytics, competencies were assessed through questionnaires, portfolios, and observations, with data analyzed using t-tests, ANCOVA, and thematic analysis. Main Findings: The experimental group demonstrated significantly higher graduate competency scores (M=81.89, SD=9.78) compared to the control group (M=70.02, SD=10.02), t(90)=5.98, p<.001, Cohen's d=1.25, indicating a large effect size. Analysis of digital portfolio data revealed enhanced self-regulated learning behaviors and deeper engagement with character development activities. Qualitative analysis showed that students appreciated the gamification elements and found the LMS-mediated learning more engaging and meaningful. Learning analytics data indicated consistent progress tracking and timely teacher interventions. Novelty/Originality of this study: The technology-enhanced Tri Hita Karana character education model effectively develops primary students’ competencies through culturally grounded digital pedagogy. This study strengthens evidence on technology-mediated character education and shows how local wisdom supports global educational goals. It highlights the need for teacher professional development and policy support to implement culturally responsive, technology-enhanced, and sustainable blended learning in primary education.

Blended Learning for Character Education: Integrating Tri Hita Karana Wisdom to Develop Graduate Competencies Aligned with SDGs Astawan, I Gede; Utama, I Dewa Gede Budi; Paramartha, Wayan Eka; Abdurahman, Ayi; Maidin, Siti Sarah
Journal of Educational Technology and Learning Creativity Vol. 3 No. 2 (2025): December
Publisher : Cahaya Ilmu Cendekia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37251/jetlc.v3i2.2498

Purpose of the study: This study examines the effectiveness of a technology-enhanced character education model based on the Balinese Tri Hita Karana philosophy in developing primary students’ competencies aligned with SDG 4. Using blended learning, it addresses gaps in research on culturally grounded technology integration in character education. Methodology: This quasi-experimental study involved 92 fifth-grade students from four primary schools in Karangasem, Bali, comparing a Technology-Enhanced Tri Hita Karana Character Education Model with conventional instruction. Using LMS, digital portfolios, gamification, and learning analytics, competencies were assessed through questionnaires, portfolios, and observations, with data analyzed using t-tests, ANCOVA, and thematic analysis. Main Findings: The experimental group demonstrated significantly higher graduate competency scores (M=81.89, SD=9.78) compared to the control group (M=70.02, SD=10.02), t(90)=5.98, p<.001, Cohen's d=1.25, indicating a large effect size. Analysis of digital portfolio data revealed enhanced self-regulated learning behaviors and deeper engagement with character development activities. Qualitative analysis showed that students appreciated the gamification elements and found the LMS-mediated learning more engaging and meaningful. Learning analytics data indicated consistent progress tracking and timely teacher interventions. Novelty/Originality of this study: The technology-enhanced Tri Hita Karana character education model effectively develops primary students’ competencies through culturally grounded digital pedagogy. This study strengthens evidence on technology-mediated character education and shows how local wisdom supports global educational goals. It highlights the need for teacher professional development and policy support to implement culturally responsive, technology-enhanced, and sustainable blended learning in primary education.

Title

Found 28 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 28 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 28 Documents
Search