WELCOME
We have developed some novel text processing techniques which have
proven extremely effective in detecting even very small amounts of plagiarism.
The Ferret software can detect copying in text documents
and in programming code;
it can quickly alert you to any pair of documents in a pile of coursework
that you need to look at more closely.
The
paper James presented at the
i-TCE and the
slides he used describe Peter's successful
prototype Web system for searching for sources of plagiarism on the Web.
The
Ferret
version 4 release
is now available (just click on this text on our Web page to
download it completely free of charge: installation is
straightforward).
This version includes automatic conversion of MS Word and pdf files.
A manual for version 4
is also available.
Version 3 of Ferret was demonstrated
at the 2006
Plagiarism Conference.
This is a much improved version which has been totally rewritten by
Peter Lane.
A paper
describing the demonstration is also available.
JunPeng Bao, a Royal Society Visiting
Fellow from Xi'an Jiaotong University in
the People's Republic of China, has done some work on incorporating
the Chinese language into Ferret.
Prelimiary work indicated that because of the way the Ferret algorithm
operates,
this would be relatively easy to do; indeed it was,
and some screen shots show the difference
between Chinese text
where some copied text has been detected
and dissimilar Chinese text.
If you have any comments on our software,
or any contributions to make regarding
the detection of plagiarism or these
webpages feel free to
drop us a line.
History
Previous versions of Ferret
are still available.
References
The web page
of Caroline Lyon
has links to some of our publications.
Context
The UH Plagiarism Detection Group is contained within
the
School
of Computer
Science in the
Faculty of Engineering and Information Sciences of the
University of Hertfordshire.
