<![CDATA[Research in the field of text mining is now increasingly being carried out because of various industries and public figures who want to get information related to public opinion about products or individual assessments obtained from social media, both opinions that are ordinary opinions and sarcasm. In the process of doing text mining, there are many classification methods that can be used, one of which is the Support Vector Machine method which can be optimized so that it can classify data into three classification classes, namely SVM One Against One and One Against Rest. The data used in the study were 2072 data from social media twitter. The results obtained from this study are the accuracy value which has the same value, whether it is done randomly or not randomly, with a value of 60.82% randomized and 60.93% non-random. On other values such as precision, recall and F1 score, the SVM One Against Rest method has a superior value compared to the SVM One Against One value.]]>
Copyrights © 2021
