Competence Classification of Twitter Users Using Support Vector Machine (SVM) Method
Muhammad Haqqi Ghufran Rifaldi, Erwin Budi Setiawan · 2019
Twitter is one of the most popular social media in Indonesia. We freely share and write ideas, information or ideas into the available column (tweet). This convenience makes twitter users have different potential and behavior. In accordance with the experience of himself or his environment. Therefore, it is necessary to have the classification of twitter users to determine and obtain the competence of whether the tweet is credible and in accordance with the condition of the twitter user. The use of the Support Vector Machine method with sequential training optimization on the classification of Twitter tweet data competency, can be used to predict the level of credibility of twitter users. Based on the working principle of the method is to determine a straight line or the best hyperplane that separates two data classes. Then the results obtained in this study are in the form of classification accuracy in 5 categories in 5 scenarios of sharing training data and different testing data, and classification of competencies in each account tested.