A Pairwise Document Analysis Approach for Monolingual Plagiarism Detection

The task of plagiarism detection entails two main steps, suspicious candidate retrieval and pairwise document similarity analysis also called detailed analysis. In this paper we focus on the second subtask. We will report our monolingual plagiarism detection system which is used to process the Persian plagiarism corpus for the task of pairwise document similarity. To retrieve plagiarised passages a plagiarism detection method based on vector space model, insensitive to context reordering, is presented. We evaluate the performance in terms of precision, recall, granularity and plagdet metrics.