Reverse engineering using association rules

Software systems need to evolve as business requirements, technology and environment change. Very often, these changes to the software are not documented, hence it becomes difficult to understand and manage such systems. To gain system understanding when documentation is non-existent or incomplete, we use reverse engineering. In this paper, we explore the use of data mining for software reverse engineering i.e. given the source files of a software system, we use association rule mining algorithms and tools to gain insight about the software. Our purpose is to determine whether association rule mining can be used for finding interesting patterns and associations within the software, that can lead to program understanding and, if required, re-structuring. We apply association rule mining to a test system and present our results. Finally we analyze our results and suggest modifications to improve the structure of the software.