Comparative Study of Existing Plagiarism Detection Systems and Design of a New Web Based System

Abstract

The purpose of this study was to compare and contrast existing plagiarism detection systems. In addition to this study, a new web based plagiarism detection system was developed. The new system uses a graph-based data-mining algorithm, SubdueGL, and a graph grammar matching algorithm, SubMatch. The attempt in the new system was to see if the graph grammar data mining technique can produce a more reliable system that can overcome some of the weaknesses discovered in the existing systems. The existing systems were found to have some limitations due to the algorithms used and system design. The new system was found to give reliable results when compared with them. The results were superior in some cases because it can detect similar patterns between documents even if there is some structural dissimilarity. The interface provided is web-based and easy to use.