A New Approach to Cross-Language Plagiarism Detection
Marc Franco-Salvador, Parth Gupta, Paolo Rosso · 2013
Cross-language variant of automatic plagiarism detection tries to detect plagiarism among documents across language pairs. In recent years a few approa- ches are proposed that use thesauri, alignment models or statistical dictionaries to deal with the similarity across languages. We propose a new approach to the cross- language plagiarism detection that makes use of a multilingual semantic network to generate knowledge graphs, obtaining a context model for each document which the other methods lack. To evaluate the proposed method, we use the Spanish-English and German-English partitions of the PAN-PC'11 corpus and compare our results with two state-of-the-art approaches. Experimental results indicate its potential to be a new alternative for similarity analysis in cross-language plagiarism detection.