Measuring the Level of Plagiarism of Thesis using Vector Space Model and Cosine Similarity Methods
Indriyanto (ac), I D Sumitra (b)
a) Postgraduate of Information System Department, Universitas Komputer Indonesia
b) Information Management Department, Universitas Komputer Indonesia
c) indriyanto[at]email.unikom.ac.id
Abstract
Plagiarism in a thesis can occur due to accidental or intentional, it is necessary to make a system that can detect plagiarism. computerized systems can help in quickly detecting and measuring the plagiarism of a scientific work. there are many techniques that can be used to measure the level of plagiarism of a document. in the papper, we will use the Vector Space Model and Cosine Similarity method to measure the degree of similarity of the thesis. the result of this method is to compare the level of thesis similarity with the dataset using the TF and TF-IDF techniques in graphical form. can be concluded from the experimental process, using the TF-IDF technique produces smaller values compared with using the TF technique
Keywords: plagiarism, scientific work, space model vector, cosine similarity.
Topic: Informatic and Information System