Document Plagiarism Detection Application Using Web-Based TF-IDF and Cosine Similarity Methods

Abstract

Listen

Translate article

In the era of technology and information which is developing very rapidly recently, this has resulted in easy access to information which makes the learning process easier in the world of education, but this ease also triggers acts of plagiarism which is a serious threat to science. Plagiarism is an act of stealing or taking someone else's work without giving proper attribution or you could say without citing that person. Therefore, an application was developed that can overcome this problem, namely a plagiarism detection application that uses the TF-IDF (Term Frequency-Inverse Document Frequency) and cosine similarity algorithm methods. TF-IDF and Cosine Similarity will be implemented into the application to carry out the calculation process which will ultimately provide results in the form of a percentage of the calculations that have been carried out. This plagiarism application is designed to detect similarities between documents in the database and user documents. The processes that occur in the application include preprocessing processes, tf-idf calculations, and cosine similarity calculations. The results of the tests carried out can be said to be consistent because the results of manual and application tests show percentage results of 4% and 4.34%. The application will also be website-based, and will be designed in such a way that it can be used to detect plagiarism.