ANALISIS SENTIMEN MASYARAKAT TERHADAP ISU RANCANGAN UNDANG-UNDANG OMNIBUS LAW PADA DATA TWITTER MENGGUNAKAN METODE SVM

Abstract

Omnibus Law is a law that is made to target one big issue that may be able to revoke or amend several laws at once so that it becomes simpler. The government's decision to combine several laws to be further simplified into an Omnibus Law has generated various responses from the public. Some of the community responses agreed or supported the results of the decision, and others did not agree. To find out the response from the public to government regulations regarding the omnibus law, Twitter is a good medium to see it through tweets from Twitter users, whether the comments given are positive, negative, or neutral comments. By classifying these comments into a data mining method, namely Support Vector Machine (SVM). The results of the application of the support vector machine method in classifying public sentiment data against the Omnibus Law Bill resulted in 5 combinations of data testing. The distribution of data of 90% : 10% produces the highest level of accuracy, which is 81%, while the distribution of data of 50% : 50% produces the lowest level of accuracy, which is 73%. Data sharing is very influential on the level of accuracy, where more and more test data can affect the results of a prediction so that it gets high accuracy