Uniqueness Control of Student Reports Based on MinHash Algorithm and Metrics for Picking Source Documents

The success of the educational process is largely due to the independence of the students' processing of the studied information. The results of this process should be in the form of a report for review and evaluation by the teacher. Unfortunately, it is quite common for students to resort to a dishonest way of preparing reports - plagiarism. Certainly, such copied papers should not be accepted by the teacher, and should be sent for reworking. However, it is far from always possible for a faculty member to quickly and easily detect the fact of plagiarism. For some types of reports, for example - graduate qualification papers, generally known plagiarism detection systems can be used, showing the percentage of uniqueness of the checked document. However, such systems are inappropriate for analyzing simpler and mass produced documents such as lab and coursework reports. To solve the problem of facilitating the work of a teacher, it is proposed to analyze ways of creating and implementing software similar to existing anti-plagiarism systems, but focused on local limited use within a set of disciplines of one teacher and providing the ability to customize the database of source documents.