A novel approach for Software Clone detection using Data Mining in Software

Abstract

The Similar Program structures which recur in variant forms in software systems are code clones. Many techniques are proposed in order to detect similar code fragments in software. The software maintenance is generally helped by maintenance is generally helped by the identification and subsequent unification. When the patterns of simple clones reoccur, it is an indication for the presence of interesting higher-level similarities. They are called as Structural Clones. The structural clones when compared to simple clones show a bigger picture of similarities. The problem of huge number of clones is alleviated by the structural clones, which are part of logical groups of simple clones. In order to understand the design of the system for better maintenance and reengineering for reuse, detection of structural clones is essential. In this paper, a technique which is useful to detect some useful types of structural clones is proposed. The novelty of the present approach comprises the formulation of the structural clone concept and the application of data mining techniques. A novel approach is useful for implementation of the proposed technique is described.