A Knowledge Based Approach to Detection of Idea Plagiarism in Online Research Publications

Plagiarism is on the rise because of the easy access to information through World Wide Web. Web pages are growing in the internet on daily basis. Researchers want to be well connected globally to popularize their ideas. Therefore, allowing download of research documents are inevitable. However, this falls prey to those who turn the cake and spoil the issue. Even unknowingly, a researcher ends in verbatim copying of other former researchers’ ideologies or conclusions to quote / use in their own research paper. This paper presents an analysis of NLP based plagiarism detection approaches which leads to proposing of an ontology based solution to detect text plagiarism more meaningfully. We address wordword and paraphrasing techniques and investigate the use of ontology in detecting idea plagiarism. The main objective is to investigate the exclusion of ‘Related Work’ section and the use of WordNet for plagiarism detection in research publications.