Plagiarism Detection in Student TextsMany researchers and academics have observed that the use of ICT has resulted in an increase in student plagiarism - and especially in web plagiarism. The availability and use of computers in teaching has seen an increase in the rate of plagiarism among students because of the wide availability of electronic texts online. Many students are tempted to copy text from electronic sources such as course materials, the work of other students, and the Internet.
Numerous styles of plagiarism exist. These range from the most common one, verbatim copying, to making use of a ghostwriter. Some of the most common forms of plagiarism are:
- "Copy-paste" verbatim copying
- Paraphrasing: the reordering of sentences and effecting changes to the words, grammar and style
- The insertion of deliberate errors: inserting spelling or grammatical mistakes into the source text
- The intentionally inaccurate use of references: the citing of incorrect or non-existent sources and the deliberate omission of sources
Although some automatic plagiarism detection systems are already available, our experimental evaluations of these systems has shown that their accuracy in detecting plagiarism leaves a lot to be desired. One of the methods of classifying detection systems is to divide them into hermetic and web detectors. In contrast to a hermetic detection system that scrutinizes only local collections of documents (other students' texts and lecture materials, for example), a web plagiarism detection system is capable of identifying instances of plagiarism from outside of the set of essays that is being analyzed. Such systems as the latter include the Internet in their ambit of their scrutiny.
AntiPlag is an advanced plagiarism detection tool intended for use on student texts. AntiPlag is one of the components of a semi-automatic essay assessment system, EssayAid, that was developed by the educational research group of the University of Joensuu in Finland. The current version of AntiPlag is a Java application that uses standard string matching techniques and web search engines in order to perform hermetic and web plagiarism detection. It is capable of detecting copying from the Internet and from local documents (such as other students' essays and course materials).