Generative AI in Computer Science Education: Insights from Topic Modeling and Text Network Analysis

The rapid evolution of generative artificial intelligence (AI) is reshaping educational landscapes., particularly in computer science education. This study investigates research trends in generative AI applications through text network and topic modeling analyses. A comprehensive literature review across IEEE Xplore., Scopus., and Web of Science identified 151 studies published between 2023 and 2024. Text network analysis revealed that “AI.,” “student.,” “education.,” “learning.,” and “performance” were the most frequently occurring terms., highlighting key research themes. Ego-network analysis demonstrated strong interconnectivity between educational AI tools and student learning outcomes. Using Latent Dirichlet Allocation (LDA)., four major research topics emerged: educational chatbots (35.1 %)., AI literacy (25.8%)., exam performance (21.8%)., and technology integration (17.2^%). Findings indicate a dominant focus on chatbots for student engagement., while ethical concerns and multimodal AI applications remain underexplored. The study underscores the necessity of addressing AI literacy gaps and enhancing interdisciplinary AI integration in educational settings. Despite its potential., generative AI adoption is hindered by privacy risks., uneven technological access., and the lack of standardized policies. Future research should focus on ethical AI frameworks., multimodal AI tools., and long-term learning outcomes. This study provides a data-driven foundation for understanding the evolving role of generative AI in education and its implications for educators., policymakers., and researchers.