Abstract:
Plagiarism is one of the growing issues in academia
and is always a concern in Universities and other academic
institutions. The situation is becoming even worse with the
availability of ample resources on the web. This paper focuses on
creating an effective and fast tool for plagiarism detection for
text based electronic assignments. Our plagiarism detection tool
named AntiPlag is developed using the tri-gram sequence
matching technique. Three sets of text based assignments were
tested by AntiPlag and the results were compared against an
existing commercial plagiarism detection tool. AntiPlag showed
better results in terms of false positives compared to the
commercial tool due to the pre-processing steps performed in
AntiPlag. In addition, to improve the detection latency, AntiPlag
applies a data clustering technique making it four times faster
than the commercial tool considered. AntiPlag could be used to
isolate plagiarized text based assignments from non-plagiarised
assignments easily. Therefore, we present AntiPlag, a fast and
effective tool for plagiarism detection on text based electronic
assignments.