A Holistic Approach to Duplicate Publication and Plagiarism Detection Using Probabilistic Ontologies

Abstract

Duplicate publication and plagiarism are two major problems in scholarly world and even they are called the cancer of academia. Plagiarism detection systems try to find similar publications of a specific article; yet, there is a little advance in holistic plagiarism detection systems. Text similarity services, without a human manual confirmation, are not capable to confirm duplication; nonetheless, it is achievable to develop a system that determines the probability of an infringement. In this paper we introduce a technique to develop such systems by using probabilistic ontologies and reasoning. The output of this system can be used for statistical surveys about rate of prevalence of plagiarism. As well, it can hit on the most probable cases of plagiarism for further investigation by human. © Springer-Verlag Berlin Heidelberg 2012.