A tree-based conceptual matching for plagiarism detection

Abstract

This paper discusses a new plagiarism detection method for text documents called Tree-based Conceptual Matching. The proposed method not only represents the content of a text document as a tree, but it also captured the underlying semantic meaning in terms of the relationships among its concepts. The method was adopted to detect plagiarism in text documents. The tree-based played a very important role in this method. It looked at the amount of detecting plagiarized sentences from the original documents. Experiments have been carried out using the CS11 standard plagiarism detection corpus. The results were evaluated using information retrieval measurements, which are Recall, Precision and F-measure. The results were compared with other methods for plagiarism detection and we found our method outperforms the other methods for plagiarism detection.