| Presentation | 2021-01-27 A change untangling technique based on learning structural features of chunks Yukitaka Sanada, Takashi Kobayashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Commits that are the result of multiple tasks, such as bug fixing and refactoring in parallel, are called compound commits. They complicate the problem of reviewing and reverting changes. In order to deal with this problem, various researches have been conducted to automatically split compound commits. However, the existing studies are based on heuristic features, and they fail to recognize complex features. In this study, we propose a method to divide commits by clustering based on the relation among chunks, which is estimated by neural networks trained with the structural features of chanks in past changes. We apply the proposed method to data from five OSS projects, and measure the accuracy of estimating the join/separation between chunks and the accuracy of commit partitioning, and compare it with previous studies. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | |
| Paper # | MSS2020-42,SS2020-27 |
| Date of Issue | 2021-01-19 (MSS, SS) |
| Conference Information | |
| Committee | MSS / SS |
|---|---|
| Conference Date | 2021/1/26(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shigemasa Takai(Osaka Univ.) / Takashi Kobayashi(Tokyo Inst. of Tech.) |
| Vice Chair | Atsuo Ozaki(Osaka Inst. of Tech.) / Kozo Okano(Shinshu Univ.) |
| Secretary | Atsuo Ozaki(Setsunan Univ.) / Kozo Okano(Hokkaido Univ.) |
| Assistant | Naoki Hayashi(Osaka Univ.) / Shinpei Ogata(Shinshu Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Mathematical Systems Science and its applications / Technical Committee on Software Science |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A change untangling technique based on learning structural features of chunks |
| Sub Title (in English) | |
| Keyword(1) | |
| Keyword(2) | |
| Keyword(3) | |
| Keyword(4) | |
| Keyword(5) | |
| 1st Author's Name | Yukitaka Sanada |
| 1st Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech.) |
| 2nd Author's Name | Takashi Kobayashi |
| 2nd Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech.) |
| Date | 2021-01-27 |
| Paper # | MSS2020-42,SS2020-27 |
| Volume (vol) | vol.120 |
| Number (no) | MSS-342,SS-343 |
| Page | pp.pp.78-83(MSS), pp.78-83(SS), |
| #Pages | 6 |
| Date of Issue | 2021-01-19 (MSS, SS) |