Plagiarism Detection System Level of Similiarity of Title Thesis Using the Winnowing Algorithm (Case Study: Information Systems Study Program)

The Winnowing Algorithm is a system built to determine the similarity of the title of a proposed thesis with previously existing thesis titles, so it is necessary to apply an algorithm to determine the similarity of titles. One of the algorithms for determining the similarity of titles is the Winnowing algorithm. The Winnowing algorithm is a method used to detect similar words/sentences (common subsequence) in two thesis titles being compared. Two thesis titles are known to have the same words / sentences if a fingerprint is found in the document, this fingerprint will be used as a basis for comparison between thesis titles, this algorithm will look for fingerprints (similarities in the two thesis titles) by changing the n-gram of the title into the form The number value is called a hash value, the technique for finding this value is hashing. The research results show that the winnowing algorithm is proven to be effective in detecting similarities in thesis titles in information systems study programs with a threshold of 10%. The system built is able to provide accurate and fast results in identifying similarities in thesis titles, and this system really helps students in finding topics that are original and have never been researched before. Apart from that, the supervisor can easily verify the similarity of the thesis title to avoid plagiarism.