User Analysis of Info BMKG Application in The Perspective of Human Computer Interaction Using Support Vector Machine Algorithm

On the Google Play Store, users often read other users' app reviews and reputations, before downloading an app. This makes analysis of user very interesting for owners to make future decisions. study aims analyze Info BMKG application on using sentiment analysis. review uses Support Vector Machine (SVM) method. The evaluation proposal was made from more than 3,000 collected INFOBMKG Store. results produce accuracy 85.54% most frequently reviewed positive are "Good", while negative "Error". Which indicates a complaint against users, words that appear often, there two combinations together, namely word "very helpful" "less accurate", which complain about problems related performance. process testing 3000 data fold = 5 test value in method obtained produces predictions testing, 1500 reviews.