plagiarism
assessment
registry
papers
software
links
discussion

PLAGUE software

We have been developing a suite of software initially for personal use. These are now successively being reimplemented or "polished" for broader, initially Monash internal, use. We are also improving and benchmarking these tools in research projects. There is always an interesting PhD, Master (Research) or Honours project in any of these areas. Contact the relevant PLAGUE members if you wish to contribute to, learn about, or use some of these tools:

Detector heuristics

Programs

Common approaches to computer program plagiarism detection include the following (see also the papers cited):
  • elimination of comments and layout
  • elimination of case
  • elimination of identifier replacement
  • elimination of program output variations
  • elimination of string/character constant variations
  • elimination of order of functions, classes
  • elimination of order of case/if-else branches
  • elimination of redundant syntax
  • compression of lexical token streams
  • reduction of artificial bias resulting from tight specifications or reusable exercise templates

General

Common patterns to be recognised by tools or to check in interviews:
  • Identical English mistakes in comments
  • Strange dates or titles in file headers, in page headers or footers, such as the name of the original author, references to non-local departments, source URLs etc
  • References outdated by several years
  • Dead referenced URLs
  • References the student does not have copies of or cannot explain
  • Inability of the student to summarise the main points of the submitted assigment
  • Inability of the student to explain critical points of the assignment in detail

 

Copyright © 2000-2002 Monash University - Credits: All Rights Reserved - Disclaimer: No warranty of integrity of these pages.
Please contact Heinz Schmidt, to submit any errors or improvement suggestions for this page.
$Id: software.shtml,v 1.9 2003/11/06 04:38:27 hws Exp $