A hybrid-token and textual based approach to find similar code segments

In this paper we propose a system capable of finding similar source code segments which are called as code clones. Such system would help maintainer solve their problem in a fast and efficient manner. It has been found that a code consist of 7 to 23 percent of cloning which make task difficult. So to solve such a problem we propose a hybrid approach of token and textual to find code cloning solving the problem encountered in tool named BOREAS which was not capable of finding type-3 types of code clones. BOREAS tool so designed was capable to match the speed that of tree matching algorithm and hence this tool would extend the performance over the present tool.