Triaging incoming change requests: Bug or commit history, or code authorship?

Abstract

-There is a tremendous wealth of code authorship information available in source code. Motivated with the presence of this information, in a number of open source projects, an approach to recommend expert developers to assist with a software change request (e.g., a bug fixes or feature) is presented. It employs a combination of an information retrieval technique and processing of the source code authorship information. The relevant source code files to the textual description of a change request are first located. The authors listed in the header comments in these files are then analyzed to arrive at a ranked list of the most suitable developers. The approach fundamentally differs from its previously reported counterparts, as it does not require software repository mining. Neither does it require training from past bugs/issues, which is often done with sophisticated techniques such as machine learning, nor mining of source code repositories, i.e., commits.An empirical study to evaluate the effectiveness of the approach on three open source systems, ArgoUML, JEdit, and MuCommander, is reported. Our approach is compared with two representative approaches: 1) using machine learning on past bug reports, and 2) based on commit logs. The presented approach is found to provide recommendation accuracies that are equivalent or better than the two compared approaches. These findings are encouraging, as it opens up a promising and orthogonal possibility of recommending developers without the need of any historical change information.

Auto-completing bug reports for Android applications

“…A subset of prior work has focused on bug and crash triage [66,58,41,45,74,46,65,15,48,53]. The techniques associated with this topic typically employ different program analysis and machine learning or natural language processing techniques to match bug reports with appropriate developers.…”

Collaborative bug triaging using textual similarities and change set analysis

“…It exposes data contained in the data model to the views of the user interface, ensuring that the architecture of the application obeys the MVVM pattern 7 . The Interactive User Interface is responsible for presenting all the data to the user and provides means to interact with the data intuitively and collaboratively.…”

“…Other approaches have extended these measures, for instance Kagdi et al [14], who also analyze source code repositories and differentiate between two types of expertise, deep and wide expertise. Researchers also used heuristics on source code history to identify potential expert developers for bug reports, e.g., Kagdi et al and Linares-Vasquez et al [8], [7] who first retrieve source code components related to a bug report using an information retrieval technique and then analyze the version control history of these components to recommend potential experts.…”

“…[6]). To determine an expert for a bug report, researchers then use information retrieval techniques to identify parts of the code that are related to a bug report and recommend the developers that are considered experts in these parts [7], [8].…”

Effective assignment and assistance to software developers and reviewers

“…Machine learning techniques and information retrieval are used in different areas such as bug triaging, impact analysis, feature location, bug severity and classifying software changes [97,58,99]. Different machine learning techniques such as ML-KNN, SVM, LSI or decision tree are also used for recommending developers to resolve a reported bug request [105,106,66].…”