MaPla: A Marathi Plagiarism Checker using Document Synset Matrix
Plagiarism is a challenging issue in the education and research domain. The plagiarism detecting tools are based on the mapping of words. Currently, very few existing tools make use of the semantic relationship between the words to detect plagiarism. Incorporating semantism in the process of detecting plagiarism gives more accurate plagiarism results. These tools work on English text, but detecting plagiarism for Natural languages like Marathi is required as text data in regional languages is growing exponentially. The proposed approach uses a synset based method called a Document synset matrix for Marathi (DSMM), a Document term matrix for Marathi (DTMM) to calculate the similarity between the documents and detects plagiarism. Also, plagiarism is tested manually and using a semantic-based tool working on translated English text. The comparison of all methods is done on originally written and rephrased documents by Marathi linguists, similarity found out for both original and rephrased documents, using DSMM is more closer to manually tested similarity measure.
The abstract is to be in fully-justified italicized text as it is here, below the author