Review of Recent Plagiarism Detection Techniques and Their Performance Comparison

Abstract

With the explosive growth of technology and the easy availability of content on the web, it creates new challenges to discriminate against the original work from plagiarized material. Content is said to be plagiarized when it is taken from other original sources without giving its reference. To address this issue Plagiarism detection tools are required. Over the years, extensive work has been done in the development of anti-plagiarism tools. This paper presents the types of plagiarism with an aim to review Extrinsic Plagiarism detection techniques using Linguistic-based features, Syntactic-based features, and Semantic-based features. Further, an overview of some current state of art methodologies and their results has been discussed on the dataset of PAN-PC 2009, PAN-PC 2010, and PAN-PC 2011. This paper also analyzes the pros and cons of some existing systems and by comparing results it also identifies that some of the systems have less potency to detect the manual and highly shuffled complex types of plagiarism such as translation obfuscation.