Predicting Personality Based on Social Media Behavior of Mobile Terminal

For hundreds of millions of people, mobile social media has become an essential part of their life. Conceivably, more and more people will be attached into this way of life. Under this background, social media on mobile terminal provides researchers with a lot of information about users' social behaviors. This paper attempts to predict users' personality with their information on social media. For 231 users from Sina Weibo, a Chinese social media, we filtered out those users whose mobile social behaviors are less than 90% of their total online behaviors. Then we conducted Big-Five personality tests with remaining 86 qualified users and predicted their personality by machine learning techniques.