We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Software Engineering

Title: Root cause prediction based on bug reports

Abstract: This paper proposes a supervised machine learning approach for predicting the root cause of a given bug report. Knowing the root cause of a bug can help developers in the debugging process - either directly or indirectly by choosing proper tool support for the debugging task. We mined 54755 closed bug reports from the issue trackers of 103 GitHub projects and applied a set of heuristics to create a benchmark consisting of 10459 reports. A subset was manually classified into three groups (semantic, memory, and concurrency) based on the bugs' root causes. Since the types of root cause are not equally distributed, a combination of keyword search and random selection was applied. Our data set for the machine learning approach consists of 369 bug reports (122 concurrency, 121 memory, and 126 semantic bugs). The bug reports are used as input to a natural language processing algorithm. We evaluated the performance of several classifiers for predicting the root causes for the given bug reports. Linear Support Vector machines achieved the highest mean precision (0.74) and recall (0.72) scores. The created bug data set and classification are publicly available.
Comments: 6 pages
Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI)
Journal reference: Proceedings of the 2020 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW), Coimbra, Portugal, 2020, pp. 171-176
DOI: 10.1109/ISSREW51248.2020.00067
Cite as: arXiv:2103.02372 [cs.SE]
  (or arXiv:2103.02372v1 [cs.SE] for this version)

Submission history

From: Thomas Hirsch [view email]
[v1] Wed, 3 Mar 2021 12:47:15 GMT (441kb,D)

Link back to: arXiv, form interface, contact.