A Low-Level Structure-based Approach for Detecting Source Code Plagiarism

Karnalim, Oscar (2017) A Low-Level Structure-based Approach for Detecting Source Code Plagiarism. IAENG International Journal of Computer Science, 44 (4). ISSN 1819-656X

[img] Text
A8 2017-12 IAENG IJCS ISSN-1819-9224.pdf - Published Version
Download (2368Kb)

Abstract

According to the fact that source code plagiarism is an emerging issue in Computer Science programming courses, several source code plagiarism detection approaches are developed. One of them is Karnalim’s approach, an approach which detects plagiarism based on low-level tokens. This paper proposes an expansion of such approach by incorporating three contributions which are: flow-based token weighting; argument removal heuristic; and invoked method removal. Flow-based token weighting aims to reduce the number of false-positive results; argument removal heuristic aims to generate moreaccurate linearized method content; and invoked method removal aims to fasten processing time. According to our evaluation, three findings can be deducted about proposed approach. Firstly, advantages provided by our proposed approach are prominent in both controlled and empirical environment. Secondly, our proposed approach outperforms Karnalim’s and state-of-the-art approach in terms of time efficiency. Finally, our approach is moderately effective to handle plagiarism attacks in practical environment.

Item Type: Article
Uncontrolled Keywords: plagiarism detection, source code, low-level language, java, bytecode
Subjects: T Technology > T Technology (General)
Depositing User: Perpustakaan Maranatha
Date Deposited: 09 Apr 2018 01:52
Last Modified: 09 Apr 2018 01:52
URI: http://repository.maranatha.edu/id/eprint/24321

Actions (login required)

View Item View Item
MCUrepository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.