A SURVEY ON CLUSTERING TECHNIQUES FOR IDENTIFICATION OF EXTRACT CLASS OPPORTUNITIES

Suchithra Chandran, Bright Gee Varghese. R

Abstract: Refactoring is a growing research area in the field of software remodularization. Refactoring is an essential practice in software development field. Refactoring is done to clean up the code and to minimize the chance of introducing the bugs. Extract class refactoring is done to improve the design of the system thereby increasing the cohesion among the class members and reducing the coupling between two classes. Extract class refactoring is performed on large, complex and less cohesive classes, which are doing functions that should be split into two or more classes. Such large and complex classes are decomposed to several classes during refactoring. During refactoring a new class is created and the entities that perform a function are moved to it. For extract class refactoring the classes to be extracted for refactoring has to be identified first. The identified refactoring opportunities are then evaluated to check whether they preserve the original behavior of the system. Refactoring is performed even after the release of the software to improve its performance. Clustering is a technique used to group the similar entities in a class. Several clustering algorithms are used to find these refactoring opportunities. In this survey various clustering techniques to identify those classes are reviewed considering its advantages and disadvantages.

Keywords: Clustering, Object-oriented system, Refactoring, Software remodularization

DOI: https://doi.org/10.15623/ijret.2013.0212071