[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127512-en":3,"doc-seo-127512-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127512,13056712833777,"Logic","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Graph-Based Machine Learning Improves Just-in-Time Defect Prediction","Increasing software complexity requires collaboration among thousands of developers, making defect-prone changes more likely to introduce faults. Identifying when such changes are introduced remains difficult, and traditional machine learning-based Just-In-Time (JIT) defect prediction has plateaued. This work constructs contribution graphs linking developers and source files, then applies graph-based ML to classify edges that represent defect-prone changes. Experiments on 14 open-source projects show substantially improved predictive quality over state-of-the-art JIT approaches, including high F1 and Matthews correlation coefficients, and the paper discusses limitations and operational use for JIT prediction.","arXiv :2110 .05371v3 [ cs . SE] 14 Apr 2023  \nGraph-based machine learning improves just-in-time defect prediction  \nJonathan Bryan 1¶‡* , Pablo Moriano2¶*  \n1 AT&T Cybersecurity, AT&T, Atlanta, GA, USA  \n2 Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA  \n¶These authors contributed equally to this work.  \n‡The contribution of this author was made when he was a Science Undergraduate Laboratory Intern at Oak Ridge National Laboratory.  \n* Corresponding author  \n[E-mail: jz699j@att.com](E-mail: jz699j@att.com) (JB), [moriano@ornl.gov](moriano@ornl.gov) (PM)  \nAbstract  \nThe increasing complexity of today's software requires the contribution of thousands of developers. This complex collaboration structure makes developers more likely to introduce defect-prone changes that lead to software faults. Determining when these defect-prone changes are introduced has proven challenging, and using traditional machine learning (ML) methods to make these determinations seems to have reached a plateau. In this work, we build contribution graphs consisting of developers and source files to capture the nuanced complexity of changes required to build software. By leveraging these contribution graphs, our research shows the potential of using graph-based ML to improve Just-In-Time (JIT) defect prediction. We hypothesize that features extracted from the contribution graphs may be better predictors of  \ndefect-prone changes than intrinsic features derived from software characteristics. We corroborate our hypothesis using graph-based ML for classifying edges that represent defect-prone changes. This new framing of the JIT defect prediction problem leads to remarkably better results. We test our approach on 14 open-source projects and show that our best model can predict whether or not a code change will lead to a defect with an F1 score as high as 77:55% and a Matthews correlation coefficient (MCC) as high as 53:16% . This represents a 152% higher F1 score and a 3% higher MCC over the state-of-the-art JIT defect prediction. We describe limitations, open challenges, and how this method can be used for operational JIT defect prediction.  \n1 Introduction  \nSoftware quality assurance, including source code inspection and testing, has become increasingly necessary for building high-quality software [1] . Software defects, or bugs, are detrimental to software quality and have a negative economic and reputational impact on software stakeholders, especially when they lead to software failures [2] . Thus, there is a huge incentive to detect likely software defects as early as possible in the development process. Reducing the number of software defects through quick and automatic identification would lead to the production of better software by improving its usability and reducing costs associated with maintenance.  \nPrevious research on software quality assurance focuses on either module-level [3] or Just-In-Time (JIT) defect prediction [4] . The module-level approach uses machine learning (ML) models trained on historical data obtained from software characteristics, including code churn, change metadata, and complexity metrics [5] . Defect prediction models detect defect-prone software modules (e.g., files [6], subsystems [7]) . Defect prediction models are then used to identify software modules that likely contain faulty code. These models can also help prioritize software quality assurance efforts, such as code reviews and pre-release testing. The JIT approach, in contrast, focuses on change-level defect prediction. This means that the focus is on software changes (i.e. , commits) rather than on modules.  \nJIT has important advantages over module-level defect prediction [4] . First, it reduces defect detection time: JIT predictions are obtained when changes are ready to  \nbe committed, before the software has been deployed. Second, it provides attribution: JIT predictions are linked to the author of the change ","cbCaimu4fTQ5PcjF","https://ap.wps.com/l/cbCaimu4fTQ5PcjF","pdf",803181,1,22,"English","en",105,"# Abstract\n# Introduction\n## Software defects and incentives for early detection\n## Module-level vs. Just-In-Time (JIT) defect prediction\n## Limitations of traditional JIT features and models\n## Contribution graphs and graph-based machine learning","[{\"question\":\"What problem does the paper address in software quality assurance?\",\"answer\":\"It addresses identifying when defect-prone changes are introduced during development, where traditional JIT defect prediction has reached limited improvement.\"},{\"question\":\"How does the proposed method differ from traditional JIT defect prediction?\",\"answer\":\"It builds contribution graphs linking developers and source files, using graph-based ML to predict defect-prone changes via edge classification rather than relying mainly on intrinsic software characteristics.\"},{\"question\":\"What results does the paper report from evaluating the approach?\",\"answer\":\"Tests on 14 open-source projects show markedly better performance than state-of-the-art JIT defect prediction, including reported F1 and MCC improvements.\"}]","Graph-Based Machine Learning Improves Just-in-Time Defect Prediction | PDF",1785939606,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"graph-based-machine-learning-improves-just-in-time-defect-prediction","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/graph-based-machine-learning-improves-just-in-time-defect-prediction/127512/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the paper address in software quality assurance?","Question",{"text":76,"@type":77},"It addresses identifying when defect-prone changes are introduced during development, where traditional JIT defect prediction has reached limited improvement.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed method differ from traditional JIT defect prediction?",{"text":81,"@type":77},"It builds contribution graphs linking developers and source files, using graph-based ML to predict defect-prone changes via edge classification rather than relying mainly on intrinsic software characteristics.",{"name":83,"@type":74,"acceptedAnswer":84},"What results does the paper report from evaluating the approach?",{"text":85,"@type":77},"Tests on 14 open-source projects show markedly better performance than state-of-the-art JIT defect prediction, including reported F1 and MCC improvements.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]