Paraphrase type identification for plagiarism detection using contexts and word embeddings

Abstract Paraphrase types have been proposed by researchers as the paraphrasing mechanisms underlying acts of plagiarism. Synonymous substitution, word reordering and insertion/deletion identified some common strategies used plagiarists. However, similarity reports generated most plagiarism detection systems provide a score produce matching sections text with their possible sources. In this research we propose methods to identify two important paraphrase – synonymous substitution in paraphrased, plagiarised sentence pairs. We three staged approach that uses context pretrained embeddings for identifying reordering. Our indicates use Smith Waterman Algorithm Plagiarism Detection ConceptNet Numberbatch produces best performance terms $$\hbox {F}_1$$ F 1 scores. This can be complement currently available incorporating detection.