PERBANDINGAN N-GRAM TECHNIQUE DAN RABIN KARP PADA APLIKASI PENDETEKSI PLAGIARISME DOKUMEN TEKS BAHASA INDONESIA

Plagiarism is copying or take over a works, paper and so on from other people who seemed to be claimed as his own work. Plagiarism detection can be done by matching the test document and source document. Detection process has few steps, which is preprocessing step, TF-IDF weighting step, and document similiarity detection process. Detection process is done with N-Gram Technique and Rabin Karp method. N-gram do the sentence splitting based on defined character length, then do the character matching and counting the similiarity percentage value. While in the Rabin Karp detection do the sentence splitting based on character length, then counting the hash value, then matching it, if hash value is match, then for the next step is doing the word matching and counting the similiarity percentage value. From the test result, TF-IDF weighting can be use to plagiarism detection. From the comparison of both method, N-Gram and Rabin Karp produce the same similarity value, but different detection time. N-Gram detection process time is faster than Rabin Karp