PhD Dissertation Proposal by YANG Chengran | From Q&A to Intelligent Software Development: Empowering Development through Data-Driven Summarization

ABOUT THE TALK

The exponential growth of technical content on Q&A platforms like Stack Overflow has created a vast, valuable resource for developers seeking insights on API usage, troubleshooting, and code optimization. However, leveraging this wealth of information effectively remains a challenge due to the variability and volume of user-generated content. This dissertation proposal aims to bridge this gap by introducing novel summarization and classification techniques that transform Q&A data into actionable insights for software development. 

The proposed research comprises three main contributions. First, we introduce TechSumBot, a summarization model that generates high-quality summaries of technical answers within the software engineering domain, significantly outperforming existing approaches in both automated and user evaluations. Second, we advance API review classification by fine-tuning pre-trained transformer models, yielding substantial improvements in categorizing community feedback into meaningful aspects. Lastly, we present APIDocBooster, a framework that combines extractive and abstractive summarization to enhance API documentation by integrating concise, relevant content from external sources.

ABOUT THE SPEAKER

Chengran YANG is a fourth-year PhD candidate in Computer Science at Singapore Management University, under the supervision of Prof David LO. He completed his bachelor's degree at the University of Electronic Science and Technology of China. His research interest is data-driven automated software engineering, particularly leveraging Q&A data to support developer tasks.