Mining and predicting micro-process patterns of issue resolution for open source software projects

• Main Authors: WANG, Yiran, CAO, Jian, LO, David
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2020
• Subjects: Issue lifetime prediction; Issue pattern prediction; Issue resolution; Micro-pattern; Process mining; Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/5628; https://ink.library.smu.edu.sg/context/sis_research/article/6631/viewcontent/Mining_predicting_micro_process_2020_31.pdf
• Institution: Singapore Management University

Addressing issue reports is an integral part of open source software (OSS) projects. Although several studies have attempted to discover the factors that affect issue resolution, few pay attention to the underlying micro-process patterns of resolution processes. Discovering these micro-patterns will help us understand the dynamics of issue resolution processes so that we can manage and improve them in better ways. Of the various types of issues, those relating to corrective maintenance account for nearly half hence resolving these issues efficiently is critical for the success of OSS projects. Therefore, we apply process mining techniques to discover the micro-patterns of resolution processes for issues relating to corrective maintenance. Four and five typical patterns are found for the identification stage and solving stage of the resolution processes respectively. Furthermore, it is shown that the consequent patterns can be predicted with a certain degree of accuracy by selecting the appropriate features and models. Furthermore, we make use of the pattern information predicted to forecast the issue lifetime and the results show that this information can also improve the accuracy in the earlier observation points. At the same time, pattern predictions provide good interpretability to the forecast of issue lifetime.