Plagiarism Detection for Programming and Natural Languages

Project Title

Plagiarism Detection for Programming and Natural Languages

Project Leader

Prof David Rossiter

School / Dept

SENG / COMP

Project Duration

Jan 2004 - Dec 2004

Project Description

This project aimed to assess a submitted document, determine its level and location of plagiarism by comparing its content with that of other documents. The proposed tool was part of the system, Mark My Words, so it could be adopted as a regular assessment item. Because it addressed plagiarism of both computer and natural languages, the tool could be applied across disciplines. The data might provide valuable insight into the level of plagiarism, and so might be employed widely in planning future educational policies.

Project Outcome

• Plagiarism detecting was supported in two modes: 1) automatic flagging of suspect text in a number of files when those files are submitted to the server, and 2) selection of particular text segments by a marker during the marking of an assignment.

• The program allowed the degree and location of plagiarism to be reliably determined. The types of plagiarism which couold be detected include  plagiarism from the web, plagiarism from other students and  self-plagiarism from previously submitted assignments

• Template was also developed which advised students how to avoid plagiarism. These were enhanced based on user feedback to provide step-by-step assistance in academic citation formats.

Status

Completed

Project Documents
(Only accessible by HKUST users)

• Project Plan
• Final Report

Adaptation

Adaptation from Mark My Words