Categorizing and Visualizing Issue Tickets to Better Understand the Features Implemented in Existing Software Systems
2019 10th International Workshop on Empirical Software Engineering in Practice (IWESEP)(2019)
摘要
Acquiring knowledge and context of the ongoing project is one of the most challenging issues for new project members. New project members need to comprehend the contents of features already implemented in software systems. However, most software documents (e.g., flow chart and data-model) were initially created and not updated. Herein we focus on tickets issued during the project period because they are created based on stakeholders' requirements as a project evolves. We propose a novel approach to categorize and visualize tickets to better understand the features implemented in the existing software systems. Specifically, tickets are grouped by a clustering method and the number of clusters (i.e., ticket categories) is automatically estimated. Then the ticket categories are visualized by (i) creating a heatmap of the tickets lifetimes, (ii) extracting keywords of the ticket categories, and (iii) analyzing the relationships between ticket categories. We applied our approach to an NTT software development project. Additionally, we interviewed the project members and external experts to evaluate the effectiveness of our approach. Our approach helps comprehend the features of the software system.
更多查看译文
关键词
reverse engineering, software comprehension, issue ticket, text mining, machine learning, interviewing
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络