Applying Data Mining on Execution Trace Log File for Improving Maintainability

Abstract

Software Engineering is a domain which has, in a short span of time, provided a vast scope for researchers. It has become an important part in software development process since people and organizations mostly rely on advanced software systems. Advanced software system requires the skills and directed efforts during the development phase and thus needs to be engineered. This has increased the competition for better software development which in turn has aroused an urgent need to emphasize on improving the software performance and quality. In our research, we apply data mining on software engineering data to enhance the maintainability of the system. The execution trace log files are used as the software engineering data. The mining algorithm identifies the most frequently accessed data from the logs. Analyzing the result of mining algorithm along with the logger levels in the log file the error prone area of the code is identified. Once the sensitive part of the code is recognized more emphasis on this part of the code would ensure minimum defects and errors in this code during various stages of software development life cycle thus improving the overall quality and performance of the software system.