Boosting Neural Commit Message Generation with Code Semantic Analysis
Wed 13 Nov 2019 14:40 - 14:55 at South Park - Student Research Competition - Selected Presentations (Undergraduate) Chair(s): Jie M. Zhang, Jin L.C. Guo
It has been long suggested that commit messages can greatly facilitate code comprehension. However, developers may not write good commit messages in practice. Neural machine translation (NMT) has been suggested to automatically generate commit messages. Despite the efforts in improve NMT algorithms, the quality of the generated commit messages is not yet satisfactory. This paper, instead of improving NMT algorithms, suggests that proper preprocessing of code changes into concise inputs is quite critical to train NMT. We approach it with semantic analysis of code changes. We collect a real-world dataset with 50k+ commits of popular Java projects, and verify our idea with comprehensive experiments. The results show that preprocessing inputs with code semantic analysis can improve NMT significantly. This work sheds light to how to apply existing DNNs designed by the machine learning community, e.g., NMT models, to complete software engineering tasks.
Tue 12 Nov
Wed 13 Nov
13:40 - 15:20: Student Research Competition - Student Research Competition - Selected Presentations (Undergraduate) at South Park Chair(s): Jie M. ZhangUniversity College London, UK, Jin L.C. GuoMcGill University | ||||||||||||||||||||||||||||||||||||||||||
13:40 - 13:55 | Crowdsourced Report Generation via Bug Screenshot Understanding Shengcheng YuNanjing University, China | |||||||||||||||||||||||||||||||||||||||||
13:55 - 14:10 | Towards Comprehensible Representation of Controllers using Machine Learning Gargi BalasubramaniamBirla Institute of Technology and Science, Pilani, K K Birla Goa Campus | |||||||||||||||||||||||||||||||||||||||||
14:10 - 14:25 | Empirical Study of Python Call Graph Li YuNanjing University | |||||||||||||||||||||||||||||||||||||||||
14:25 - 14:40 | A Machine Learning based Approach to Identify SQL Injection Vulnerabilities Kevin ZhangWayne State University | |||||||||||||||||||||||||||||||||||||||||
14:40 - 14:55 | Boosting Neural Commit Message Generation with Code Semantic Analysis Shuyao JiangFudan University |