Plagiarism Detection for Indonesian Texts

Abstract

In this paper, we describe a view of our research method on the Plagiarism Detection for Indonesian texts that we are working on. This method should address the problems of handling the equivalence class of Indonesian tokens, selecting the targeted source documents, and minimizing the gap of similarity measurement between the selection and the comparison modules. For these reasons, we propose a novel document representation in the candidate document retrieval module and the hybrid of segmentation and similarity of hashing technique in the comparison module.