Towards Auto-labelling Issue Reports for Pull-Based Software Development using Text Mining Approach

Fazayeli, Hassan; Syed-Mohamad, Sharifah Mashita; Akhir, Nur Shazwani Md.

doi:10.1016/j.procs.2019.11.160

Cited by 10 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study differs from the previous studies [1], [2], [4], [18]- [23], [26], [27] since our study focuses on the usage of multi-labels. Our study goes one step further than previous studies [1], [2] by showing that the projects with multi-labels have a shorter resolution time than the projects without multilabels.…”

Section: B Studies On Issue Labeling and Classificationmentioning

confidence: 88%

“…Researchers have proposed techniques to classify issue reports into two categories: bugs and non-bugs [18]- [25]. In addition, Kochhar et al proposed classifying issue reports into thirteen pre-defined categories [26], and Fazayeli et al proposed classifying issue reports into five categories [27]. Recently, Kallis et al proposed an automatic labeling system, called Ticket-Tagger [4].…”

Section: B Studies On Issue Labeling and Classificationmentioning

confidence: 99%

See 1 more Smart Citation

An Empirical Study on Using Multi-Labels for Issues in GitHub

Kim

Lee

2021

IEEE Access

View full text Add to dashboard Cite

On GitHub, one of the most successful services for software project hosting, labels have been used to represent various information and decisions about reported issues. However, previous studies on labels were limited to simple statistics or label recommendations for issue types. In this paper, we aim to provide a better understanding of labels and their usage in software development. We particularly focus on using multiple and custom labels on issues. To analyze label usage, we collected software project data and label usage information from GitHub. We then quantitatively investigated the performance of projects with multi-label features and qualitatively investigated the categories of multi-labels, and the usage of multilabels based on these categories. Our analysis results show that multi-labels are common in the majority of software projects and that projects using multi-label features manage their issues more effectively. In addition, our analysis results reveal different types of information represented by labels, which are related to features, development, and issues. This study finds several facts that can be used for studies on issue management and thus that help develop labeling techniques to mitigate the burden of issue management.

show abstract

Section: B Studies On Issue Labeling and Classificationmentioning

confidence: 88%

Section: B Studies On Issue Labeling and Classificationmentioning

confidence: 99%

An Empirical Study on Using Multi-Labels for Issues in GitHub

Kim

Lee

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Fazayeli et al [5] used text-based classification to label GitHub issues. However, they only evaluated their framework for only a single repository i.e.…”

Section: Related Workmentioning

confidence: 99%

“…To create an issue report 5 in the GitHub issue tracker, users are required to provide a title and a brief description as the body. However, the labels associated with the issue are optional.…”

Section: Introductionmentioning

confidence: 99%

Automatic Issue Classifier: A Transfer Learning Framework for Classifying Issue Reports

Nadeem,

Sarwar,

Malik

2022

Preprint

View full text Add to dashboard Cite

Issue tracking systems are used in the software industry for the facilitation of maintenance activities that keep the software robust and up to date with ever-changing industry requirements. Usually, users report issues that can be categorized into different labels such as bug reports, enhancement requests, and questions related to the software. Most of the issue tracking systems make the labelling of these issue reports optional for the issue submitter, which leads to a large number of unlabeled issue reports. In this paper, we present a state-of-the-art method to classify the issue reports into their respective categories i.e. bug, enhancement, and question. This is a challenging task because of the common use of informal language in the issue reports. Existing studies use traditional natural language processing approaches adopting key-word based features, which fail to incorporate the contextual relationship between words and therefore result in a high rate of false positives and false negatives. Moreover, previous works utilize a uni-label approach to classify the issue reports however, in reality, an issue-submitter can tag one issue report with more than one label at a time. This paper presents our approach to classify the issue reports in a multi-label setting. We use an off-the-shelf neural network called RoBERTa and fine-tune it to classify the issue reports. We validate our approach on issue reports belonging to numerous industrial projects from GitHub. We were able to achieve promising F-1 scores of 81%, 74%, and 80% for bug reports, enhancements, and questions, respectively. We also develop an industry tool called Automatic Issue Classifier (AIC), which automatically assigns labels to newly reported issues on GitHub repositories with high accuracy.

show abstract