| Presentation | 2021-03-03 Performance Evaluation of Automatic Bug Repair using Neural Machine Translation with Bug Fix Histories Gakuto Akiyama, Tsukasa Nakamura, Yasutaka Kamei, Naoyasu Ubayashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | DeepFix is one of the automatic bug fixing tools. It repairs syntax errors by embedding errors into error-free programs and applying machine translation technology to learn how to generate the programs before embedding errors from the programs with the embedded errors. However, it is difficult to repair errors other than the embedded errors, and the previous studies have shown that the fixed-rate decreases when there is a difference between the errors in the target data and the embedded errors. Therefore, we focused on Learning-Fixes. It learns the repairs using machine translation from buggy programs and fixed programs produced by the developers and automatically repairs bugs. Although the previous studies of Learning-Fixes have focused on semantic errors, we considered that Learning-Fixes could be applied to syntax errors as well because it learns the repairs with sequence-to-sequence. In this study, we evaluated the performance of Learning-Fixes on syntax errors by applying Learning-Fixes to the data used in the performance evaluation of DeepFix, which was said to have a lower fixed-rate. As a result, typical syntax errors were repaired with a fixed-rate of 20-54 %, and the result showed that Learning-Fixes has sufficient performance for syntax errors. Furthermore, we showed that Learning-Fixes was able to repair typographical errors, unlike DeepFix. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Automatic bug repair / Machine Translation |
| Paper # | SS2020-34 |
| Date of Issue | 2021-02-24 (SS) |
| Conference Information | |
| Committee | SS |
|---|---|
| Conference Date | 2021/3/3(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Takashi Kobayashi(Tokyo Inst. of Tech.) |
| Vice Chair | Kozo Okano(Shinshu Univ.) |
| Secretary | Kozo Okano(Hiroshima City Univ.) |
| Assistant | Shinpei Ogata(Shinshu Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Software Science |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Performance Evaluation of Automatic Bug Repair using Neural Machine Translation with Bug Fix Histories |
| Sub Title (in English) | |
| Keyword(1) | Automatic bug repair |
| Keyword(2) | Machine Translation |
| 1st Author's Name | Gakuto Akiyama |
| 1st Author's Affiliation | Kyushu University(Kyushu Univ.) |
| 2nd Author's Name | Tsukasa Nakamura |
| 2nd Author's Affiliation | Kyushu University(Kyushu Univ.) |
| 3rd Author's Name | Yasutaka Kamei |
| 3rd Author's Affiliation | Kyushu University(Kyushu Univ.) |
| 4th Author's Name | Naoyasu Ubayashi |
| 4th Author's Affiliation | Kyushu University(Kyushu Univ.) |
| Date | 2021-03-03 |
| Paper # | SS2020-34 |
| Volume (vol) | vol.120 |
| Number (no) | SS-407 |
| Page | pp.pp.37-42(SS), |
| #Pages | 6 |
| Date of Issue | 2021-02-24 (SS) |