An Anti-Plagiarism Add-on For WebCAT

Plagiarism is a major problem in every discipline and Computer Science courses are no different. It is very common for students to submit their peers programming assignment as their own. This practice is unfair and also halts the learning process of the students who choose to copy. This research investigates the performance of various software plagiarism detection tools such as MOSS, JPlag and Plaggie. Controlled changes were made to a code file, and the sensitiveness of the various tools to those changes was determined. Plaggie with its algorithm of tokenisation followed by string comparison was found to have acceptable performance for our tests. It is also open source whereas the other tools are proprietary and web based and we decided to incorporate Plaggie with Web-CAT. Web-CAT is a flexible, automated grading system designed to process computer programming assignments. It serves as a learning environment for software testing tasks and helps automatically assess student assignments. The developed system was tested using submissions to a real class assignment and also by a variety of potential future users in a number of tests and the feedback received was very positive. In addition the tests presented a number of possible future enhancements