Research on the Model of Similar Electronic Coursework Detection

摘要

With rapid development of network technologies, many Universities have realized coursework electronically in many courses. But students copy electronic coursework more and more commonly. This paper proposes two methods of anti-plagiarism plagiarism both from Internet and among students: If the source of plagiarism is from Internet, we adopt the technology based on the semantic tree similarity detection, in which distributed database and parallel computation are used. If the source of plagiarism is among students, we use a new way based on digital watermarking and information hiding technology. Our program can embed the information of originators into the characters of coursework by the watermarking algorithm with high embedding capacity. When the coursework is copied by others, the program can detect the similar parts and locate the source of plagiarism accurately.