Towards the taxonomy of Structural Clones by using data Mining techniques

Cloning refers to the duplication of code in the software system. Software developers prefer cloning as they are under tight time constraints and also for the performance reasons. Redundant code also has negative impacts in software maintenance & evaluation. Copying of code can also indicate copying of bugs in the software. Simple Clones are formed by similar text fragment but there are large program structures which are formed by the configuration of simple clones, called structural clones. Earlier various techniques have been proposed for detecting similar code fragments in the software, which are called simple clones. But we can obtain further gains by elevating the level of code clone analysis. Recurring patterns of simple clones indicates the presence of higher level similarities i.e. structural clones In this paper, we will propose a technique to detect some useful types of structural clones. This approach will detect structural clones by using data mining technique and clustering etc. It uses the follow up technique on simple clones for finding high level clones. We will describe a tool called clone miner that implements this technique.