Cross-language similarity estimation is relevant for different cross-language tasks such as information retrieval, clustering, and categorisation, among others. In this research, we are interested in the automatic detection of cross-language text reuse and plagiarism.

Models for cross-language text similarity estimation are often complicated and require different resources, such as thesauri, dictionaries and bilingual corpora. In this research work we aim to simplify the problem by approaching it on the basis of the cognateness convept, proposed by Simard et al. (1992)