Monolingual and Crosslingual Plagiarism Detection Towards the Competition @ SEPLN 09 ⋆

Automatic plagiarism detection considering a reference corpus compares a suspicious text to a set of documents in order to relate the plagiarised fragments to their potential source. The suspicious and source documents can be written wether in the same language (monolingual) or in different languages (crosslingual). In the context of the Ph. D., our work has been focused on both monolingual and crosslingual plagiarism detection. The monolingual approach is based on a search space reduction process followed by an exhaustive word n-grams comparison. Surprisingly it seems that the application of the reduction process has not been explored in this task previously. The crosslingual one is based on the well known IBM-1 alignment model. Having a competition on these topics will make our work available to the Spanish scientific community interested in plagiarism detection.