Since students turn in hundreds or even thousands of programs every year,
the efficiency of the system is a very important factor to reduce the workload of
the markers. We aim to provide a system that is able to find out the plagiarized
programs within a very short period of time. Table 2 shows the efficiency of our
system. To test 100 programs with an average 100 lines, the system needs to
make 4950 comparisons and it took only 10 seconds to complete all the
comparisons.
The system was tested by colleagues in the computing team in The Open
University of Hong Kong. It has been tested and used by tutors in some
programming courses offered in distance learning mode. The system was
further enhanced to have the following features:
1. Allowing markers to run the plagiarism detection part in a batch mode
2. Capable of checking and opening the compressed files (*.zip)
3. Filtering out those pairs with thresholds
4. Outputting the testing reports to users in html format
Conclusions
Plagiarism is currently a serious problem in academic institutions. Programming
assignments, especially computer programs, can easily be copied and modified
230 S. C. Ng, S. O. Choy and R. Kwan
by students. An accurate and flexible system can help markers identify the
plagiarized pairs of program among a large set of programs. This chapter reports
on the development of an integrated online system with assignment submission,
management and marking system for handling programming assignments in
distance learning courses. The system can streamline the process of marking
programming assignments and relieve the workload of instructors and markers.
As a result, students can enjoy the benefits of online submission. With features
like plagiarism detection, automatic program testing and the online marking
sheet, the system enables the teaching staff to mark and manage assignments
more efficiently and accurately.
Acknowledgments
The authors would like to thank the OUHK and Education and Manpower
Bureau (EMB) of the Hong Kong SAR Government for the Earmarked Research
Grant (Ref: 9004/03H) that allowed the system to be developed.