<![CDATA[YouTube is the world's largest online video social media that is used to display various videos created by users and companies in the field of media content. Every video contained in YouTube can be done by giving a text type comment in the comments column of the video that has been watched. The large number of comments causes the content creator (video maker) to spend enough time to understand every emotion in the existing comment. After consideration of the solution used to resolve the problem, the authors chose to use the Modified K-Nearest Neighbor (MKNN) classification method with BM25 and Chi-Square feature selection. The test used is 5-fold cross validation to find the best k value which is then used for testing the Chi-Square feature selection. In Chi-Square test the data used is the best fold data based on the highest f-measure value in the 5-fold cross validation test. The results obtained are the maximum accuracy, precision, recall, f-measure values ​​achieved when k is 30, 72,82%, 72,94%, 72,26%, and 72,59%. While the Chi-Square test on the 4th fold of data the best number of terms used is 40% and 50%, with the value of accuracy, precision, recall, f-measure is 80,56%, 80,37%, 81,61 % and 80,98%.]]>
Copyrights © 2020
