Deep Learning-based Production and Test Bug Report Classification using Source Files
Classifying production and test bug reports can significantly improve not only the accuracy of performance evaluation but also the performance of information retrieval-based bug localization (IRBL). However, it is time-consuming for developers to classify these bug reports manually. As an alternative, deep learning offers promising solutions for classifying bug reports automatically. This study proposes a production and test bug report classification method based on deep learning. Our method uses a set of source files and model tuning to solve the problem of insufficient and sparse bug reports when applying deep learning. Our experimental results reveal that the macro f1-score of our method is 34% better than that of the classifier trained with only bug reports and can improve the mean average precision of VSM-based IRBL by 22%.
Wed 11 MayDisplayed time zone: Eastern Time (US & Canada) change
03:00 - 04:00 | |||
03:00 5mPoster | A Static Analyzer for Detecting Tensor Shape Errors in Deep Neural Network Training Code Posters Ho Young Jhoo Seoul National University, Sehoon Kim Seoul National University, Woosung Song Seoul National University, Kyuyeon Park Seoul National University, DongKwon Lee Seoul National University, South Korea, Kwangkeun Yi Seoul National University, South Korea Pre-print | ||
03:05 5mPoster | Garuda: Heap aware symbolic execution Posters | ||
03:10 5mPoster | The Symptoms, Causes, and Repairs of Workarounds in Apache Issue Trackers Posters Aoyang Yan Shanghai Jiao Tong University, Hao Zhong Shanghai Jiao Tong University, Daohan Song Shanghai Jiao Tong University, Li Jia Shanghai Jiao Tong University | ||
03:15 5mPoster | CrystalBLEU: Precisely and Efficiently Measuring the Similarity of Code Posters | ||
03:20 5mPoster | CRISCE: Towards Generating Test Cases from Accident Sketches Posters Vuong Nguyen University of Passau, Alessio Gambi University of Passau, Jasim Ahmed University of Passau, Gordon Fraser University of Passau | ||
03:25 5mPoster | Deep Learning-based Production and Test Bug Report Classification using Source Files Posters Misoo Kim Sungkyunkwan University, Youngkyoung Kim Sungkyunkwan University, Eunseok Lee Sungkyunkwan University | ||