AN ITERATIVE GENETIC ALGORITHM BASED SOURCE CODE PLAGIARISM DETECTION APPROACH USING NCRR SIMILARITY MEASURE

摘要

With the advent of WWW and improvements in globally accessible software warehouses attained that source code is enthusiastically reachable to software designers. Even though, reusing of source code has its individual benefits, precaution is to be taken to guarantee that patented software [19] does not invade any authorizations. In this Context, Plagiarism Detection plays a very significant role. Although several existing detection approaches has been introduced, most of the approaches work on one to many similarity measures. However, this might not be very much helpful in case large number of datasets where many-to-many relationship exist. In this paper, an intelligent detection model is purposed by employing the iterative genetic algorithm with two different fitness evaluation functions. Prior to the detection model, the source code is preprocessed to remove noise and dimensionality reduction techniques are employed. The experimental results for the proposed approach are carried out using two different data sets. From the experimental results, it is found that the proposed model has good performance compared to the other existing approaches such as fuzzy clustering based Detection system and Incremental Genetic Algorithm.