Document Similarity Detection using K-Means and Cosine Distance

A two-year study by the Ministry of Research, Technology and Education in Indonesia presented the evaluation of most universities in Indonesia. The findings of the evaluation are the peculiarities of various dissertation softcopies of doctoral students which are similar to any texts available on internet. The suspected plagiarism behavior has a negative effect on both students and faculty members. The main reason behind this behavior is the lack of standardized awareness among faculty members with regard to plagiarism. Therefore, this study proposes a computerized system that is able to detect plagiarism information by using K-means and cosine distance algorithm. The process starts from preprocessing process that includes a novel step of checking Indonesian big dictionary, vector space model design, and the combined calculation of K-means and cosine distance from 17 documents as test data. The result of this study generally shows that the documents have detection accuracy of 93.33%.