KLASIFIKASI PENGGUNA MEDIA SOSIAL DI KALANGAN PELAJAR SMK N 1 SANGATTA UTARA --

ABSTRACT In this modern era, we are certainly no stranger to social media. Almost all ages today use various kinds of social media, including teenagers. In the North Sangatta State Vocational High School (SMK N) 1, social media is an option to communicate to stay connected with others and know the current situation. In addition to the positive impact of using social media, there are also negative impacts, one of which is addiction to disrupt the learning process. For this reason, it is necessary to classify students of North Sangatta State Vocational High School (SMK N) 1 to determine that these students are addicted to using social media with application categories such as Instagram, WhatsApp, Facebook and Twitter. The sample data used were 317 data obtained based on questionnaires that had been distributed to students of North Sangatta State 1 Vocational High School. Then the test data used as many as 190 data for later data mining process. In this study, it is known that the Naive Bayes method is the best used compared to the other two methods, namely K-Nearest Neighbors and Decision Tree. The stages of the Naive Bayes method are counting the number of classes/labels, counting the number of cases per class, multiplying all class variables and then comparing the results per class. The results of this test, all data was successfully classified correctly and has an accuracy value of 100% and an AUC value of 1,000. Key codes: Social media, data mining, classification, Naïve Bayes, K-Nearest Neighbors, Decision Tree, sample data.