In this paper, we adopt a template matching method to detect design patterns from a software system by calculating their normalized cross correlation. Because design patterns document flexible design ideas, there can be various ways of implementing them. In our approach, not only the exact matches of pattern instances are detected from system source code, but also the variations of pattern candidates can be identified. Based on our method, we provide tool support and perform experiments on different large open-source systems.

Design pattern describes a recurring problem and its common solution, which usually is in abstract form. The application of design pattern requires coding the generic solution. It is necessary to assure the coding process correctly implements not only the structure but also the desired behavior of the design pattern. This problem is called implementation correctness in this paper. By providing the definition of partial order between sequence diagrams, we formally describe the implementation correctness. We verify the implementation correctness with model checking by using process algebra to specify the source code and temporal logic to specify the behavior of the pattern.

Recovering design patterns applied in a system can help refactoring the system. Machine learning algorithms have been successfully applied in mining data patterns. However, one of the main obstacles of applying them for design pattern detection is the difficulty of collecting training examples. Unlike other applications, a design pattern instance typically includes a group of classes with certain relationships. Thus, the possible combinations of the group of classes can be enormous which results in huge training sets making the application of machine learning algorithms impracticable. In this paper, we propose an innovative method using matrix transformations to cluster the training examples. Our method can significantly reduce the size of training examples, thus making it possible to be efficiently applied in machine learning algorithm.