engleski

Source-code preprocess model for improved plagiarism detectionin student programming assignments

Plagiarism is a big concern in programming courses. Students can copy-paste source-code parts, change it and present it as their own. It is hard to find plagiarized assignments manually, especially in large student groups and with many possible obfuscation methods. Many tools are built to automatically check for plagiarism. Since no algorithm for calculating similarity that the tools use is perfect, there are various preprocessing techniques that can help improve the quality of detection. In this research the idea is to verify how different preprocessing techniques affect different tool and define new preprocess model to improve the quality of detection.

plagiarism detection ; programming assigements ; preprocessing

nije evidentirano