<![CDATA[Human needs of technology and the availability of adequate infrastructure are evidences that the technology currently becomes part of the human beings’ basic necessities. Growing multitude of journals and scientific papers makes choosing and sorting become more selective though there have been many online journals service providers and portals. Research on search engines, plagiarism and recommendation system has been carried out with various methods to improve the performance of the system itself, this paper aims to calculate similarities between one article with other articles by implementing n-gram and tanimoto cosine. The number of articles tested were forty-three titles and abstracts, tests were carried fifty times with random selected keywords, by separating each sentences of the title and abstract into n characters (n = 2) including spaces and punctuation, then calculating similarity to the query or keywords used to test the system. Testing was done using several variation of the thresholds. After observing the fifty times-testings, the threshold value of 0.30, produced accuracy = 0.86, precision = 0.37 and recall = 0.44.]]>
Copyrights © 2018
