A clustering approach for the identification of object-oriented design patterns

Maintenance is undoubtedly the most effort consuming activity in the software process. Representing a system as a collection of high level building blocks, such as design patterns, can decrease maintenance effort. In this paper, an approach for design pattern recovery is presented. ne approach is based on clustering or grouping the classes, which contribute to a pattern using the structural attributes of design patterns. We present a grouping algorithm, which is tailored according to, the specific pattern being searched for. The advantage of this tailoring is that it reduces the complexity of the clustering algorithm. The suggested algorithm is tested on several case studies and the results are compared with other design pattern recovery algorithms. Results of the case studies demonstrate the efficiency and accuracy of the clustering algorithm.