Dark Triad Personality Traits Prediction through Twitter Analysis using Support Vector Machine Method

Personality is a combination of a person's characteristics and behaviors in any particular condition. Several personality models are used to predict personality tendencies, one of them is a personality with a dark side, or it is called the Dark Triad Personality, which is divided into three traits, namely Machiavellianism (manipulative), Narcissism (narcissistic), and Psychopath (antisocial). This study uses Dirty Dozen questionnaire data distributed publicly using Google Form as the initial label of Dark Triad personality traits. It compares with the result data from Twitter which is crawled with Twitter API using Rstudio. Twitter users can interact with each other and express themselves through tweets, so their personalities can be known through the words or characters represented in a tweet. This study uses the Support Vector Machine method to predict Dark Triad Personality tendencies through Twitter analysis. From the results of this study, the classification with the highest accuracy is 73.3%, obtained from scenario 2 with a comparison of training data and testing data is 80:20.