DEEPLINK: Recovering issue-commit links based on deep learning
DEEPLINK: Recovering issue-commit links based on deep learning
复制标题
DEEPLINK：基于深度学习恢复问题提交链接
DOI：
10.1016/j.jss.2019.110406
复制
发表时间：
2019-12
影响因子：
3.5
通讯作者：
Zhao Wenyun
中科院分区：
文献类型：
--
作者：
Ruan Hang;Chen Bihuan;Peng Xin;Zhao Wenyun
问题跟踪系统中的问题与解决版本控制系统中的问题的提交之间的链接对于各种软件工程任务（例如，错误预测、错误定位和特征定位）。然而，只有一小部分这样的链接是通过手动将问题标识符包括在提交日志中来建立的，使得它们中的大部分在演化历史中丢失。为了恢复问题-提交链接，基于知识的和基于学习的技术利用问题和提交中的元数据和文本/代码相似性;然而，它们未能捕获问题和提交中的嵌入语义以及问题和提交之间隐藏的语义相关性。因此，这种语义差距抑制了链接恢复的准确性。为了弥合这一差距，我们提出了一种语义增强的链接恢复方法，命名为DeepLink，它是建立在深度学习技术之上的。具体来说，我们开发了一个神经网络架构，使用词嵌入和递归神经网络，学习问题和提交中的自然语言描述和代码的语义表示以及问题和提交之间的语义相关性。在实验中，为了量化缺失问题提交链接的普遍性，我们分析了1078个高星级的GitHub Java项目（即，583，795个已关闭的问题），发现只有42.2%的问题与相应的提交相关联。为了评估DeepLink的有效性，我们使用10个GitHub Java项目将DeepLink与最先进的链路恢复方法FRLink进行了比较，并证明DeepLink在F度量方面优于FRLink。
The links between issues in an issue-tracking system and commits resolving the issues in a version control system are important for a variety of software engineering tasks (e.g., bug prediction, bug localization and feature location). However, only a small portion of such links are established by manually including issue identifiers in commit logs, leaving a large portion of them lost in the evolution history. To recover issue-commit links, heuristic-based and learning-based techniques leverage the metadata and text/code similarity in issues and commits; however, they fail to capture the embedded semantics in issues and commits and the hidden semantic correlations between issues and commits. As a result, this semantic gap inhibits the accuracy of link recovery.To bridge this gap, we propose a semantically-enhanced link recovery approach, namedDeepLink, which is built on top of deep learning techniques. Specifically, we develop a neural network architecture, using word embedding and recurrent neural network, to learn the semantic representation of natural language descriptions and code in issues and commits as well as the semantic correlation between issues and commits. In experiments, to quantify the prevalence of missing issue-commit links, we analyzed 1078 highly-starred GitHub Java projects (i.e., 583,795 closed issues) and found that only 42.2% of issues were linked to corresponding commits. To evaluate the effectiveness ofDeepLink, we comparedDeepLinkwith a state-of-the-art link recovery approach FRLink using ten GitHub Java projects and demonstrated thatDeepLinkcan outperform FRLink in terms ofF-measure.
登录
查看更多内容
DOI：
10.1109/csmr.2013.19
发表时间：
2013-03
期刊：
2013 17th European Conference on Software Maintenance and Reengineering
影响因子：
--
作者：
Tegawendé F. Bissyandé;Ferdian Thung;Shaowei Wang;D. Lo;Lingxiao Jiang;Laurent Réveillère
通讯作者：
Tegawendé F. Bissyandé;Ferdian Thung;Shaowei Wang;D. Lo;Lingxiao Jiang;Laurent Réveillère
DOI：
10.1145/2593882.2593891
发表时间：
2014-05
期刊：
Future of Software Engineering Proceedings
影响因子：
--
作者：
J. Cleland-Huang;O. Gotel;J. Hayes;Patrick Mäder;A. Zisman
通讯作者：
J. Cleland-Huang;O. Gotel;J. Hayes;Patrick Mäder;A. Zisman
DOI：
10.1109/wcre.2003.1287240
发表时间：
2003-11
期刊：
10th Working Conference on Reverse Engineering, 2003. WCRE 2003. Proceedings.
影响因子：
--
作者：
M. Fischer;M. Pinzger;H. Gall
通讯作者：
M. Fischer;M. Pinzger;H. Gall
DOI：
10.18653/v1/p17-2045
发表时间：
2017-04
期刊：
ArXiv
影响因子：
--
作者：
Pablo Loyola;Edison Marrese-Taylor;Y. Matsuo
通讯作者：
Pablo Loyola;Edison Marrese-Taylor;Y. Matsuo
影响因子：
7.4
作者：
Antoniol, G;Canfora, G;Merlo, E
通讯作者：
Merlo, E