Happy Coding: Software Quality Prediction using Emotional Estimation

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00096
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Software
• Research Institution: Kyoto Institute of Technology
• Principal Investigator: Mizuno Osamu 京都工芸繊維大学, 情報工学・人間科学系, 教授 (60314407)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: ソフトウェア不具合予測 / 感情推定 / リポジトリマイニング

Outline of Final Research Achievements

In this research, we analyze the relationship between the transitions of the developers' emotional polarities of commit comments and the characteristics of the projects, by applying sentiment analysis to the commit comments stored in source code repositories.
We apply this experiment to four open source software projects. As a result, we found that there are cases where there is a significant difference in their emotional polarities among those two groups of the four projects which divided based on the characteristics of the projects and mainly used programming language. We also found and analyzed that there are comments that inconvenience when applying sentiment analysis to commit comments.

Free Research Field

ソフトウェア工学