Multilingual Idea plagiarism detection for scientific text based on Word Net Dataset

Plagiarism occurs when the content is copied without any permission or citation. By increasing the scientific text, the plagiarism in this domain has been increased. This paper introduced the plagiarism detection method that recognized the plagiarism based on WordNet dataset in thirty-four different languages. In a scientific text, the proposed method works locally and used bag of words file. In this case the processing time can be improved. In addition, acceptable precision, recall and f-measure value in provided method has been showed by experimental results on PAN2014 and open multilingual WordNet dataset for thirty-four languages. So it can be suggested for scientific text and it is not limited by one language.

Keywords:

plagiarism detection , open multilingual WordNet dataset , bag of words file