Issue-Labeler: an ALBERT-based Jira Plugin for Issue Classification

Waleed Alhindi, Abdulrahman Aleid, Ilyes Jenhani, Mohamed Wiem Mkaouer
{"title":"Issue-Labeler: an ALBERT-based Jira Plugin for Issue Classification","authors":"Waleed Alhindi, Abdulrahman Aleid, Ilyes Jenhani, Mohamed Wiem Mkaouer","doi":"10.1109/mobilsoft59058.2023.00012","DOIUrl":null,"url":null,"abstract":"Issue labels are key drivers in software maintenance as they dictate the prioritization, organization, and ultimately the resolution of encountered issues. Consequently, mislabeling issues result in inefficient prioritization, which compromises the resolution process of these issues. Thus, to increase the accuracy and effectiveness of issue labeling in software maintenance, this paper proposes \"Issue-Labeler\": an automated issue labeler plugin for Jira1, which utilizes a deep neural language model to predict an issue’s type based on its title and description. Specifically, the plugin would classify an issue into three types: BUG, IMPROVEMENT, and NEW FEATURE. The issue-labeler plugin was implemented by fine-tuning Google’s pre-trained ALBERT language model, using 35,889 labeled issue reports extracted from 77 projects. The plugin showed an average F1-score of 0.75, 0.58, and 0.67, respectively, for the BUG, IMPROVEMENT, and NEW FEATURE issues. The plugin will provide developers with a tool that recommends issue labels to, in turn, optimize the process of tagging and resolving these issues. Video of tool setup and runtime is available: https://voutu.be/mi2FwaXNrR4. Tool Webpage: https://issue-labeler.github.io/issue-labeler-site/. Replication package: https://github.com/issue-labeler/.","PeriodicalId":311618,"journal":{"name":"2023 IEEE/ACM 10th International Conference on Mobile Software Engineering and Systems (MOBILESoft)","volume":"199 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 IEEE/ACM 10th International Conference on Mobile Software Engineering and Systems (MOBILESoft)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/mobilsoft59058.2023.00012","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Issue labels are key drivers in software maintenance as they dictate the prioritization, organization, and ultimately the resolution of encountered issues. Consequently, mislabeling issues result in inefficient prioritization, which compromises the resolution process of these issues. Thus, to increase the accuracy and effectiveness of issue labeling in software maintenance, this paper proposes "Issue-Labeler": an automated issue labeler plugin for Jira1, which utilizes a deep neural language model to predict an issue’s type based on its title and description. Specifically, the plugin would classify an issue into three types: BUG, IMPROVEMENT, and NEW FEATURE. The issue-labeler plugin was implemented by fine-tuning Google’s pre-trained ALBERT language model, using 35,889 labeled issue reports extracted from 77 projects. The plugin showed an average F1-score of 0.75, 0.58, and 0.67, respectively, for the BUG, IMPROVEMENT, and NEW FEATURE issues. The plugin will provide developers with a tool that recommends issue labels to, in turn, optimize the process of tagging and resolving these issues. Video of tool setup and runtime is available: https://voutu.be/mi2FwaXNrR4. Tool Webpage: https://issue-labeler.github.io/issue-labeler-site/. Replication package: https://github.com/issue-labeler/.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Issue- labeler:一个基于albert的Jira问题分类插件
问题标签是软件维护中的关键驱动因素,因为它们规定了优先级、组织,并最终解决遇到的问题。因此,错误标记问题会导致低效的优先级排序,从而损害这些问题的解决过程。因此,为了提高问题标注在软件维护中的准确性和有效性,本文提出了“issue - labeler”:Jira1的自动问题标注插件,它利用深度神经语言模型根据标题和描述来预测问题的类型。具体来说,插件会将问题分为三种类型:BUG、IMPROVEMENT和NEW FEATURE。问题标记器插件是通过微调谷歌的预训练ALBERT语言模型实现的,使用了从77个项目中提取的35,889个标记问题报告。该插件在BUG、IMPROVEMENT和NEW FEATURE问题上的平均f1得分分别为0.75、0.58和0.67。该插件将为开发人员提供一个工具,推荐问题标签,从而优化标记和解决这些问题的过程。工具设置和运行的视频可在:https://voutu.be/mi2FwaXNrR4。工具页面:https://issue-labeler.github.io/issue-labeler-site/。复制包:https://github.com/issue-labeler/。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Issue-Labeler: an ALBERT-based Jira Plugin for Issue Classification Native vs Web Apps: Comparing the Energy Consumption and Performance of Android Apps and their Web Counterparts Vulnerability Propagation in Package Managers Used in iOS Development Understanding the Impact of Fingerprinting in Android Hybrid Apps Reducing the Impact of Breaking Changes to Web Service Clients During Web API Evolution
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1