Developing an AI-Enabled Personalized Learning System for Computer Science Education

Thesis Abstract

The increasing prevalence of digital technology in education necessitates innovative approaches to enhance learning outcomes in computer science, especially through personalized instructional methods supported by artificial intelligence (AI). Despite the proliferation of e-learning platforms, many struggle to address diverse learner needs effectively, highlighting a significant gap in adaptive learning systems that tailor content to individual students' cognitive styles, prior knowledge, and learning paces. This study aims to develop and evaluate an AI-enabled personalized learning system specifically designed for computer science education, with the overarching goal of improving student engagement, comprehension, and performance in computing courses. The specific objectives include (1) identifying core components and features essential to AI-driven personalization in computer science education; (2) designing and implementing a prototype intelligent learning system incorporating machine learning algorithms for real-time learner analysis; (3) assessing the system’s usability, effectiveness, and impact on students’ academic performance; and (4) providing evidence-based recommendations for integration of AI-driven personalized learning models in higher education institutions. The research employs a mixed-methods approach, integrating quantitative experimentation with qualitative feedback to ensure comprehensive evaluation of the system. The population comprises 200 undergraduate computer science students enrolled in intermediate programming courses at a large public university. A stratified random sampling technique is used to select 100 students for the experimental group, which interacts with the AI-enabled system, and 100 students in the control group, who receive conventional instruction. Data collection instruments include pre- and post-intervention assessments, system usability questionnaires, and semi-structured interviews. Performance metrics such as course grades, completion rates, and engagement logs are collected to quantitatively evaluate learning outcomes. The validity and reliability of the quantitative instruments are established via Cronbach’s alpha and content validation by domain experts. Qualitative data from interviews are analyzed through thematic analysis, allowing for contextual insights into user experiences. Data analysis employs descriptive statistics to summarize student performance and engagement patterns, while inferential statistics such as t-tests and ANOVA determine the significance of differences between groups. Machine learning models, including decision trees and neural networks, are utilized to analyze learner data and refine personalization features iteratively. The theoretical underpinning references Vygotsky’s Social Constructivism, emphasizing the importance of scaffolding in learning, and the Adaptive Learning Theory, which supports tailoring instruction to individual learner differences. Expected findings suggest that students interacting with the AI-enabled personalized system will demonstrate significantly higher performance, increased engagement, and perceived improved learning experiences compared to traditional instruction. The system’s adaptive features are anticipated to facilitate more efficient mastery of complex computer science topics by providing customized feedback and resource recommendations. The research is expected to contribute to the body of knowledge by providing empirical evidence on the efficacy of AI in personalized learning environments within higher education, highlighting critical success factors, challenges, and opportunities for scalable deployment. The study concludes that integrating AI-driven personalization into computer science curricula can significantly enhance educational outcomes and student satisfaction. It recommends further development of adaptive algorithms, integration with existing learning management systems, and large-scale trials across diverse academic contexts. Future research avenues include longitudinal studies to examine long-term effects and the exploration of additional AI techniques such as reinforcement learning to further optimize personalization strategies. Overall, this research seeks to advance the understanding and practical application of intelligent systems in computer science education, fostering innovative instructional practices aligned with contemporary digital trends in education technology.