This research develops a Virtual Care Psychoeducation Framework to support lifestyle changes and medication adherence among individuals with coronary heart disease in Malaysia. By integrating digital health tools and behavioral interventions, the framework aims to enhance patient engagement, self-management, and long-term health outcomes. The study provides insights for healthcare providers to improve remote patient care, ultimately reducing complications and improving quality of life for affected individuals.
This research proposes a Self-Adaptive Differential Evolution approach that utilizes an ensemble of control parameters to enhance optimization performance. By dynamically adjusting parameters based on population diversity and problem characteristics, the method improves convergence speed, robustness, and solution quality. The study contributes to advancing evolutionary algorithms, making them more efficient and adaptable for solving complex real-world optimization problems.
This study develops an advanced profiling framework integrating intelligence and genomics data to personalize childhood learning. Using AI and data analytics, it identifies cognitive strengths, learning preferences, and developmental needs to enhance tailored educational strategies. By combining neuroscience, AI-driven data processing, and educational psychology, the research creates a comprehensive system to guide educators and parents in making informed learning interventions. Its findings aim to support national education policies, promoting inclusivity and optimizing resources for diverse learning abilities.
The research introduces a novel dynamic-length chromosome formulation for identifying marker genes in cancer classification. It leverages evolutionary optimization techniques to refine gene expression profiling for more accurate biomarker selection. By dynamically adjusting chromosome length, the method enhances the adaptability and efficiency of genetic algorithms. This approach improves cancer classification accuracy by selecting the most relevant genes while reducing computational complexity. Ultimately, the study contributes to advancing precision medicine through optimized genetic feature selection.
This research develops a learning sentiment model that analyzes instant messenger data to enhance teaching and learning. By leveraging natural language processing (NLP) and machine learning, the model detects student emotions, engagement levels, and learning challenges in real time. The study aims to provide educators with insights for personalized interventions, fostering a more adaptive and responsive learning environment.
Emotion Detection System for Online Learning
This research leverages AI and facial recognition to analyze student emotions during virtual classes. It detects expressions such as happiness, confusion, or frustration to assess engagement levels. Educators can use this real-time emotional feedback to adjust their teaching methods accordingly. This system enhances student participation and creates a more interactive learning environment. Ultimately, it aims to improve learning outcomes in online education.
A Game-Based Learning Analytic Platform for Assessing Children's Logic Development
This research develops a game-based learning analytics platform to assess children's logic development through interactive gameplay. By integrating AI-driven analytics, the platform tracks cognitive patterns, problem-solving skills, and learning progress in real time. The study aims to provide educators with data-driven insights to personalize instruction, enhance critical thinking skills, and improve early childhood education through engaging, gamified learning experiences.
This research develops an AI-powered learning assistant to enhance mathematics lessons by providing personalized support, real-time feedback, and interactive problem-solving guidance. By leveraging machine learning and adaptive learning techniques, the tool aims to improve student engagement, comprehension, and performance. The study contributes to AI-driven education, offering insights into effective classroom integration and personalized learning experiences to support both educators and students.
This research explores learning-based ultrasound image quality characterization to improve fetal growth assessment accuracy. By leveraging machine learning, it aims to enhance image quality evaluation, reducing variability in measurements and improving diagnostic reliability. The study supports obstetricians in making precise clinical decisions, leading to better maternal-fetal health outcomes. Its findings contribute to advancing ultrasound imaging standards and optimizing prenatal care.
This research explores how personality traits influence engagement and motivation in educational gaming. It aims to develop a personalized persuasive design that adapts to different child personality types. By incorporating psychological and behavioral factors, the framework enhances learning experiences and encourages positive behavior. The study emphasizes the importance of tailored game mechanics to improve effectiveness in digital education. Ultimately, this research contributes to the development of more engaging and impactful learning technologies for children.
Investigating Machine Behaviour Through the Computational Red Teaming Framework
This research explores machine behavior using a Computational Red Teaming Framework to assess AI systems' robustness, biases, and vulnerabilities. By simulating adversarial scenarios and stress-testing models, it aims to uncover weaknesses and improve AI reliability, security, and ethical alignment. The study provides insights into AI risk management, contributing to safer, more accountable, and resilient machine intelligence systems.