GITG: Automatic Issue Title Generation Method for Gitee Platform

YANG Jun; LIU Shi-fan; CHEN Xiang; CUI Zhan-qi

doi:10.12263/DZXB.20240434

您当前的位置：

首页 >

文章列表页 >

GITG: Automatic Issue Title Generation Method for Gitee Platform

PAPERS | 更新时间：2025-08-18

- GITG: Automatic Issue Title Generation Method for Gitee Platform
- ACTA ELECTRONICA SINICA Vol. 53, Issue 5, Pages: 1559-1570(2025)
- 作者机构：
  
  1.北京信息科技大学计算机学院，北京 100192
  2.南通大学信息科学技术学院，江苏南通 226019
- 作者简介：
- 基金信息：
  
  Jiangsu Provincial Frontier Leading Technology Fundamental Research Project(BK20202001);Beijing Information Science and Technology University “Qin-Xin Talent” Cultivation Project(QXTCPB202406)
- DOI：10.12263/DZXB.20240434
  CLC： TP311.5
- Received：10 May 2024，
  
  Revised：2025-02-21，
  
  Published：25 May 2025
- 稿件说明：
移动端阅览
杨君, 刘诗凡, 陈翔, 等. GITG：面向Gitee平台的issue标题自动生成方法[J]. 电子学报, 2025, 53(05): 1559-1570.

YANG Jun, LIU Shi-fan, CHEN Xiang, et al. GITG: Automatic Issue Title Generation Method for Gitee Platform[J]. Acta Electronica Sinica, 2025, 53(05): 1559-1570.
杨君, 刘诗凡, 陈翔, 等. GITG：面向Gitee平台的issue标题自动生成方法[J]. 电子学报, 2025, 53(05): 1559-1570. DOI：10.12263/DZXB.20240434

YANG Jun, LIU Shi-fan, CHEN Xiang, et al. GITG: Automatic Issue Title Generation Method for Gitee Platform[J]. Acta Electronica Sinica, 2025, 53(05): 1559-1570. DOI：10.12263/DZXB.20240434

摘要

在开源软件和开源平台中，开发人员可以通过提交issue来记录所发现的软件错误或提出新功能需求.由于缺乏经验、专业水平有限等原因，用户可能无法对issue内容进行准确有效地总结，导致issue标题质量较低，进而降低issue的解决效率.此外，现有的issue标题自动生成方法主要面向GitHub等英文开源平台，当应用在Gitee等国产开源平台时表现不佳.同时，现有方法主要使用issue主体描述作为输入，忽略了issue中的代码片段等重要信息.为此，本文提出一种面向Gitee平台的issue标题自动生成方法GITG（Gitee Issue Title Generation），针对包含中文和英文文本的issue，使用构建的Gitee issue数据集对支持中文的预训练模型Chinese BART（Bidirectional and Auto-Regressive Transformers）进行微调，利用issue主体描述和代码片段的双模态信息来自动生成issue标题.为验证GITG的有效性，构建了包含18 242个Gitee issue样本的数据集.实验结果表明，GITG在ROUGE-1、ROUGE-2和ROUGE-L指标上相较于iTAPE和iTiger分别至少提升了13.09%、10.18%和12.84%，在BLEU和METEOR指标上同样取得了性能提升.人工评价结果表明，GITG生成标题的平均得分在整体分数、流畅性、信息性和简洁性4个评价指标上相较iTAPE和iTiger分别至少提升了26.7%、20.8%、24.2%和20.0%.

Abstract

In open-source software and platforms

developers can submit issues to report software bugs or suggest new feature requests. Due to the lack of experience and limited professional skills

users may struggle to summarize the content of issues accurately and effectively

resulting in low-quality issue titles

which in turn decreases the efficiency of addressing issues. Additionally

existing automatic issue title generation methods are primarily designed for English open-source platforms

such as GitHub

and the performance are degraded when applied to Chinese open-source platforms

like Gitee. Furthermore

existing methods mainly use the issue body description as inputs

ignoring the code snippets in the issue. In this paper

we propose a method called GITG (Gitee Issue Title Generation) specifically designed for Gitee

an open-source platform. GITG addresses the challenge of generating issue titles for both Chinese and English text by fine-tuning the Chinese BART (Bidirectional and Auto-Regressive Transformers) pre-trained model on a constructed Gitee issue dataset. It leverages the bi-modal information from the issue body description and code snippets to generate informative and accurate issue titles. A dataset consisting of 18 242 Gitee issue samples is constructed to validate the effectiveness of GITG. Experimental results demonstrate that GITG outperforms iTAPE and iTiger by at least 13.09%

10.18%

and 12.84% on the ROUGE-1

ROUGE-2

and ROUGE-L metrics

respectively. GITG also achieves improvements in BLEU and METEOR metrics. Human evaluation results also indicate that the average scores of the titles generated by GITG are improved by at least 26.7%

20.8%

24.2%

and 20.0% in overall score

fluency

informativeness

and conciseness

respectively

compared to iTAPE and iTiger.

关键词

Keywords

references

ERFANI JOORABCHI M , MIRZAAGHAEI M , MESBAH A . Works for me! Characterizing non-reproducible bug reports [C ] // Proceedings of the 11th Working Conference on Mining Software Repositories . New York : ACM , 2014 : 62 - 71 .

SOLTANI M , HERMANS F , BÄCK T . The significance of bug report elements [J ] . Empirical Software Engineering , 2020 , 25 ( 6 ): 5255 - 5294 .

MA X X , KEUNG J W , YU X , et al . AttSum: A deep attention-based summarization model for bug report title generation [J ] . IEEE Transactions on Reliability , 2023 , 72 ( 4 ): 1663 - 1677 .

CHEN S Q , XIE X Y , YIN B G , et al . Stay professional and efficient: Automatically generate titles for your bug reports [C ] // Proceedings of the 35th IEEE/ACM International Conference on Automated Software Engineering . New York : ACM , 2020 : 385 - 397 .

LIN H , CHEN X , CHEN X J , et al . TitleGen-FL: Quality prediction-based filter for automated issue title generati-on [J ] . Journal of Systems and Software , 2023 , 195 : 111513 .

ZHANG T , IRSAN I C , THUNG F , et al . iTiger: An automatic issue title generation tool [C ] // Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering . New York : ACM , 2022 : 1637 - 1641 .

中华人民共和国工业和信息化部 . “十四五”软件和信息技术服务业发展规划 [EB/OL ] . ( 2021-11-15 )[ 2024-05-10 ] . https://www.miit.gov.cn/jgsj/ghs/zlygh/art/2022/art_f43c0 68acfb14f15b8daf4238945deb0.html https://www.miit.gov.cn/jgsj/ghs/zlygh/art/2022/art_f43c068acfb14f15b8daf4238945deb0.html .

ZHANG F J , YU X , KEUNG J , et al . Improving stack overflow question title generation with copying enhanced CodeBERT model and bi-modal information [J ] . Information and Software Technology , 2022 , 148 : 106922 .

ZHOU Y L , YANG S Y , CHEN X , et al . QTC4SO: Automatic question title completion for stack overflow [C ] // Proceedings of the 31st IEEE/ACM International Conference on Program Comprehension . Piscataway : IEEE , 2023 : 1 - 12 .

SHAO Y F , GENG Z C , LIU Y T , et al . CPT: A pre-trained unbalanced transformer for both Chinese language understanding and generation [J ] . Science China (Information Sciences) , 2024 , 67 ( 5 ): 152102 .

陈翔 , 于池 , 杨光 , 等 . 基于双重信息检索的Bash代码注释生成方法 [J ] . 软件学报 , 2023 , 34 ( 3 ): 1310 - 1329 .

CHEN X , YU C , YANG G , et al . Bash code comment generation method based on dual information retrieval [J ] . Journal of Software , 2023 , 34 ( 3 ): 1310 - 1329 . (in Chinese)

郭丹 , 姚沈涛 , 王辉 , 等 . 嵌入局部聚类描述符的视频问答Transformer模型 [J ] . 计算机学报 , 2023 , 46 ( 4 ): 671 - 689 .

GUO D , YAO S T , WANG H , et al . Embedding VLAD in transformer for video question answering [J ] . Chinese Journal of Computers , 2023 , 46 ( 4 ): 671 - 689 . (in Chinese)

孙锐 , 谢瑞瑞 , 张磊 , 等 . 基于灾难性遗忘及组合叠加擦除的跨模态行人重识别预训练方法 [J ] . 电子学报 , 2023 , 51 ( 10 ): 2925 - 2935 .

SUN R , XIE R R , ZHANG L , et al . Cross-modal pedestrian re-identification pre-training method based on catastrophic forgetting and combination superimposed eras-ure [J ] . Acta Electronica Sinica , 2023 , 51 ( 10 ): 2925 - 2935 . (in Chinese)

LIU K , CHEN X , CHEN C Y , et al . Automated question title reformulation by mining modification logs from stack overflow [J ] . IEEE Transactions on Software Engineering , 2023 , 49 ( 9 ): 4390 - 4410 .

LIU S Q , GAO C Y , CHEN S , et al . ATOM: Commit message generation based on abstract syntax tree and hybrid ranking [J ] . IEEE Transactions on Software Engineering , 2022 , 48 ( 5 ): 1800 - 1817 .

刘诗凡 , 崔展齐 , 陈翔 , 等 . MMCUP: 融合多模态信息的代码注释自动更新方法 [J ] . 计算机学报 , 2024 , 47 ( 1 ): 172 - 189 .

LIU S F , CUI Z Q , CHEN X , et al . MMCUP: Updating code comments based on multi-modal information [J ] . Chinese Journal of Computers , 2024 , 47 ( 1 ): 172 - 189 . (in Chinese)

BAJAJ K , PATTABIRAMAN K , MESBAH A . Mining questions asked by web developers [C ] // Proceedings of the 11th Working Conference on Mining Software Repositories . New York : ACM 2014 : 112 - 121 .

LIN C Y . ROUGE: A package for automatic evaluation of summaries [J ] . Text Summarization Branches Out , 2004 : 74 - 81 .

PAPINENI K , ROUKOS S , WARD T , et al . BLEU: A method for automatic evaluation of machine translat-ion [C ] // Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics . Stroudsburg : ACL , 2002 : 311 - 318 .

BANERJEE S , LAVIE A . METEOR: An automatic metric for MT evaluation with improved correlation with human judgments [C ] // Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization . Stroudsburg : ACL , 2005 : 65 - 72 .

陈翔 , 杨光 , 崔展齐 , 等 . 代码注释自动生成方法综述 [J ] . 软件学报 , 2021 , 32 ( 7 ): 2118 - 2141 .

CHEN X , YANG G , CUI Z Q , et al . Survey of state-of-the-art automatic code comment generation [J ] . Journal of Software , 2021 , 32 ( 7 ): 2118 - 2141 . (in Chinese)

YANG G , CHEN X , ZHOU Y L , et al . DualSC: Automatic generation and summarization of shellcode via transformer and dual learning [C ] // Proceedings of the 29th IEEE International Conference on Software Analysis, Evolution and Reengineering . Piscataway : IEEE , 2022 : 361 - 372 .

CARO T M , ROPER R , YOUNG M , et al . Inter-observer reliability [J ] . Behaviour , 1979 , 69 ( 3 ): 303 - 315 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Method-Level Software Bug Localization Enhanced with File Information

A Survey of Using Formal Concept Analysis for Software Maintenance

Related Author

BO Li-li

ZHU Cheng

LI Bin

SUN Xiao-bing

WEN Wan-zhi

LI Bi-xin

LI Yun

SUN Xiao-bing

Related Institution

School of Information Engineering， Yangzhou University

Key Laboratory of Safety-Critical Software （Nanjing University of Aeronautics and Astronautics）， Ministry of Industry and Information Technology

School of Information Engineering, Yangzhou University

State Key Laboratory for Novel Software Technology, Nanjing University

School of Computer Science and Engineering, Southeast University

⁰