Three Stage Transformation for Software Clone Detection

Abstract

Software engineers often build new procedures by cloning, copying an existing one with similar requirements, and slightly modifying it. While this may be easier than extracting the common part, and sharing it in a library, it increases the system size and often leads to higher maintenance costs. The occurrence of clones is highly dependent on the system architecture and development model, and has been studied in the past for a few large procedural systems. In the last decade, several researchers have investigated techniques to detect duplicated code in programs exceeding hundreds of thousands lines of code. All of these techniques have known merits and deficiencies, but as of today, little is known on where to fit these techniques into the software maintenance process. Current work proposes a three stage transformation for software clone detection. This technique makes use of substring matching and derives partially from token based technique for clone detection. Ambiguity problems in string matching algorithm are reduced by the dictionary replacement for similar keywords.