A Soft Alignment Model for Bug Deduplication
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
A Soft Alignment Model for Bug Deduplication. / Rodrigues, Irving Muller; Aloise, Daniel; Fernandes, Eraldo Rezende et al.
2020 IEEE/ACM 17th International Conference on Mining Software Repositories: MSR 2020, Proceedings; Seoul, Republic of Korea 29-30 June 2020. New York : Association for Computing Machinery, Inc, 2020. p. 43-53.Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - A Soft Alignment Model for Bug Deduplication
AU - Rodrigues, Irving Muller
AU - Aloise, Daniel
AU - Fernandes, Eraldo Rezende
AU - Dagenais, Michel
PY - 2020/6/29
Y1 - 2020/6/29
N2 - Bug tracking systems (BTS) are widely used in software projects. An important task in such systems consists of identifying duplicate bug reports, i.e., distinct reports related to the same software issue. For several reasons, reporting bugs that have already been reported is quite frequent, making their manual triage impractical in large BTSs. In this paper, we present a novel deep learning network based on soft-attention alignment to improve duplicate bug report detection. For a given pair of possibly duplicate reports, the attention mechanism computes interdependent representations for each report, which is more powerful than previous approaches. We evaluate our model on four well-known datasets derived from BTSs of four popular open-source projects. Our evaluation is based on a ranking-based metric, which is more realistic than decision-making metrics used in many previous works. Achieved results demonstrate that our model outperforms state-of-the-art systems and strong baselines in different scenarios. Finally, an ablation study is performed to confirm that the proposed architecture improves the duplicate bug reports detection.
AB - Bug tracking systems (BTS) are widely used in software projects. An important task in such systems consists of identifying duplicate bug reports, i.e., distinct reports related to the same software issue. For several reasons, reporting bugs that have already been reported is quite frequent, making their manual triage impractical in large BTSs. In this paper, we present a novel deep learning network based on soft-attention alignment to improve duplicate bug report detection. For a given pair of possibly duplicate reports, the attention mechanism computes interdependent representations for each report, which is more powerful than previous approaches. We evaluate our model on four well-known datasets derived from BTSs of four popular open-source projects. Our evaluation is based on a ranking-based metric, which is more realistic than decision-making metrics used in many previous works. Achieved results demonstrate that our model outperforms state-of-the-art systems and strong baselines in different scenarios. Finally, an ablation study is performed to confirm that the proposed architecture improves the duplicate bug reports detection.
KW - Attention Mechanism
KW - Bug Tracking Systems
KW - Deep Learning
KW - Duplicate Bug Report Detection
KW - Business informatics
UR - http://www.scopus.com/inward/record.url?scp=85093703740&partnerID=8YFLogxK
U2 - 10.1145/3379597.3387470
DO - 10.1145/3379597.3387470
M3 - Article in conference proceedings
AN - SCOPUS:85093703740
SP - 43
EP - 53
BT - 2020 IEEE/ACM 17th International Conference on Mining Software Repositories
PB - Association for Computing Machinery, Inc
CY - New York
T2 - 17th IEEE/ACM International Conference on Mining Software Repositories, MSR 2020, co-located with the 42nd International Conference on Software Engineering. ICSE 2020
Y2 - 29 June 2020 through 30 June 2020
ER -