Customising a Copying-Identifier for Biomedical Science Student Reports: Comparing Simple and Smart Analyses

摘要

The aim of our project is to develop a system for detecting student copying in Biomedical Science laboratory practical reports. We compare contrasting approaches: "simple" methods Zipping, based on a standard file-compression tool, and Bigrams, a basic comparison of frequent bigrams; and "smart" methods using commercial-strength plagiarism-checking systems Turnitin, Copycatch, and Copyfind. Both approaches successfully flag examples of copying in our Test Corpus of 218 student courseworks, but Copycatch provides more user-friendly batch-processing mechanism. Human experts go beyond word-based pattern matching, and take account of knowledge specific to our domain: methods and questions can legitimately be copied, whereas originality is more important in the Discussion section.