KIISE Transactions on Computing Practices (정보과학회 컴퓨팅의 실제 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Predicting Bug Severity by utilizing Topic Model and Bug Report Meta-Field

토픽 모델과 버그 리포트 메타 필드를 이용한 버그 심각도 예측 방법

  • 양근석 (서울시립대학교 컴퓨터과학부) ;
  • 이병정 (서울시립대학교 컴퓨터과학부)
  • Received : 2015.05.04
  • Accepted : 2015.07.07
  • Published : 2015.09.15
⟨ Previous Next ⟩

Abstract

Recently developed software systems have many components, and their complexity is thus increasing. Last year, about 375 bug reports in one day were reported to a software repository in Eclipse and Mozilla open source projects. With so many bug reports submitted, developers' time and efforts have increased unnecessarily. Since the bug severity is manually determined by quality assurance, project manager or other developers in the general bug fixing process, it is biased to them. They might also make a mistake on the manual decision because of the large number of bug reports. Therefore, in this study, we propose an approach of bug severity prediction to solve these problems. First, we find similar topics within a new bug report and reduce the candidate reports of the topic by using the meta field of the bug report. Next, we train the reduced reports by applying Naive Bayes Multinomial. Finally, we predict the severity of the new bug report. We compare our approach with other prediction algorithms by using bug reports in open source projects. The results show that our approach better predicts bug severity than other algorithms.

최근 개발된 소프트웨어들은 많은 수의 컴포넌트들을 가지고 있으며, 복잡성 또한 증가하고 있다. 지난 해 오픈소스 프로젝트 (Eclipse, Mozilla)에서는 하루에 약 375건의 버그 리포트가 제출되었다. 이렇게 증가된 버그 리포트들로 인해 개발자들의 시간과 노력이 불필요하게 증가하고 있다. 또 버그 심각도는 품질 보증 담당자, 프로젝트 매니저 또는 개발자에 의해 직접 판단되므로 그들에 의해 주관적으로 결정된다. 또한 많은 수의 버그 리포트 때문에 심각도 판단에서 실수할 수도 있다. 따라서 본 논문에서는 버그 심각도 예측 방법을 제안한다. 먼저, 새로운 버그 리포트가 제출되면, 유사한 토픽을 찾아내고 버그 리포트의 메타 필드를 이용하여 후보 버그 리포트의 범위를 줄인다. 추출된 버그 리포트를 Naive Bayes Multinomial 기법에 훈련하여 새로운 버그 리포트의 심각성을 예측한다. 오픈소스 프로젝트에 본 방법을 적용하여 본 방법이 버그 심각도 예측에 효과적이라는 것을 보인다.

Keywords

Acknowledgement

Supported by : 한국연구재단

References

  1. G. Jeong, S. Kim and T. Zimmermann, "Improving Bug Triage with Bug Tossing Graphs," Proc. of Joint Meeting of the European Software Engineering Conference and the ACM SIGSOFT Symposium on the Foundations of Software Engineering, pp. 111-120, 2009.
  2. N. Jalbert and W. Weimer, "Automated Duplicate Detection for Bug Tracking Systems," Proc. of International Conference on Dependable System & Networks, pp. 52-61, 2008.
  3. J. Xuan, H. Jiang, Z. Ren, J. Yan and Z. Luo, "Automatic Bug Triage using Semi-Supervised Text Classification," Proc. of International Conference on Software Engineering and Knowledge Engineering, pp. 209-214, 2010.
  4. Y. Tian, D. Lo and C. Sun, "Information Retrieval based Nearest Neighbor Classification for Fine-Grained Bug Severity Prediction," Proc. of Working Conference on Reverse Engineering, pp. 215-224, 2012.
  5. A. Lamkanfi, S. Demeyer, Q. D. Soetens and T. Verdonck, "Comparing Mining Algorithms for Predicting the Severity of a Reported Bug," Proc. of European Conference on Software Maintenance and Reengineering, pp. 249-258, 2011.
  6. T. Menzies and A. Marcus, "Automated Severity Assessment of Software Defect Reports," Proc. of International Conference on Software Maintenance, pp. 346-355, 2008.
  7. G. Yang, T. Zhang and B. Lee, "Towards Semi-Automatic Bug Triage and Severity Prediction based on Topic Model and Multi-Feature of Bug Reports," Proc. of Annual International Computers, Software and Applications Conference, pp. 97-106, 2014.
  8. N.K.S. Roy and B. Rossi, "Towards an Improvement of Bug Severity Classification," Proc. of 40th EUROMICRO Conference on Software Engineering and Advanced Applications (SEAA), pp. 269-276, 2014.
  9. The stanford natural language processing group, "Stanford topic modelling toolbox," in http://nlp.stanford.edu/software/tmt/tmt-0.4/, 2009.
  10. C. Friedman, P. Kra, H. Yu, M. Krauthammer and A. Rzhetsky, "GENIES: a natural-language processing system for the extraction of molecular pathways from journal articles," in Bioinformatics, Vol. 17, pp. S74-S82, 2001. https://doi.org/10.1093/bioinformatics/17.suppl_1.S74
  11. R. Kohavi, "A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection," Proc. of Joint Conference on Artificial Intelligence, pp. 1137-1143, 1995.