DNAP: Detection of News Article Plagiarism
Lu Lu, Zhou Li · 2021
With the rapid development of we media, enormous news articles are produced online, where plagiarism becomes easier. Hence, article plagiarism detection has become more and more important. Besides, existing state-of-the-art article plagiarism detector suffers from several limitations in fine grained plagiarism detection. In this paper, we propose a tool DNAP that can detect plagiarism at word-level granularity. In our approach, token based fingerprinting is designed for matching and asymmetric similarity coefficient is used for measuring similarity. We thoroughly evaluate DNAP by both mutation framework and real-world datasets. Compared with existing state-of-the-art tool, all experiment results demonstrate that DNAP is the best performing article plagiarism detection tool, shown as better recall and F1-score with high precision.