Bilingual plagiarism detector

Internet has become primary medium for information access, commerce in today´s globalized world and almost every information is available in the Internet either in the native language of the user or in a non-native language. Therefore, it becomes easier to use another author´s contents from the Internet without proper citation or reference and this tendency is increasing day-by-day. Such use of another author´s contents, thoughts, ideas, or expressions and the representation of them as one´s own original work is known as plagiarism. Though plagiarism can be found in almost every field, it is a major problem in academic area as plagiarism destroys individual´s creativity and originality and defeats the purpose of education. At present many commercial and noncommercial plagiarism detection software are available. However, most of them are unilingual in nature and none of them considers checking of Bangla documents for plagiarism. In this paper, we have introduced statistical method and method based on individual content for detecting plagiarism from English and Bangla electronic documents. The first method performs different statistical analysis of the documents for plagiarism detection whereas the second method is based on the analysis of individual contents of the documents. The system can perform plagiarism checking in a Bangla document from English documents and vice versa. It can also detect plagiarism from the documents of the same language. The system has been evaluated by real documents. We have found that our system can detect plagiarism from documents of two different languages efficiently.