Practical guidelines for change recommendation using association rule mining

Association rule mining is an unsupervised learning technique that infers relationships among items in a data set. This technique has been successfully used to analyze a system's change history and uncover evolutionary coupling between system artifacts. Evolutionary coupling can, in turn, be used to recommend artifacts that are potentially affected by a given set of changes to the system. In general, the quality of such recommendations is affected by (1) the values selected for various parameters of the mining algorithm, (2) characteristics of the set of changes used to derive a recommendation, and (3) characteristics of the system's change history for which recommendations are generated. In this paper, we empirically investigate the extent to which certain choices for these factors affect change recommendation. Specifically, we conduct a series of systematic experiments on the change histories of two large industrial systems and eight large open source systems, in which we control the size of the change set for which to derive a recommendation, the measure used to assess the strength of the evolutionary coupling, and the maximum size of historical changes taken into account when inferring these couplings. We use the results from our study to derive a number of practical guidelines for applying association rule mining for change recommendation.