研究目的
Investigating the effectiveness of micro interaction metrics (MIMs) in improving software defect prediction accuracy by leveraging developer interaction information.
研究成果
The study demonstrates that MIMs significantly improve defect prediction accuracy when combined with existing metrics, offering a cost-effective solution for software quality assurance. MIMs also provide intuitive feedback to developers, helping them recognize inefficient behaviors during software development.
研究不足
The study is limited by the availability and completeness of Mylyn data, as developers may not always submit their task context. Additionally, the systems and developers referenced may not be representative of all software development environments.
1：Experimental Design and Method Selection:
The study employs Mylyn, an Eclipse plug-in, to capture and store developer interactions such as file editing and browsing events. These interactions are used to propose micro interaction metrics (MIMs) for defect prediction.
2：Sample Selection and Data Sources:
The study uses Mylyn task session logs from Eclipse Bugzilla between December 2005 and June 2010, involving 5,973 task sessions.
3：List of Experimental Equipment and Materials:
Mylyn for capturing developer interactions, Eclipse IDE for development environment, and Weka for machine learning algorithms.
4：Experimental Procedures and Operational Workflow:
The study involves collecting Mylyn data, extracting MIMs, CMs, and HMs, building prediction models using machine learning algorithms, and evaluating the models' performance.
5：Data Analysis Methods:
The study uses F-measure for evaluating prediction accuracy, gain ratio for assessing the predictive power of individual metrics, and Wilcoxon rank-sum test for statistical hypothesis testing.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
