Abstract:-With the development of social networks, a large variety of approaches havebeen developed to define users’ personalities based on their social activitiesand language use habits. Particular approaches differ with regard to differentmachine learning algorithms, data sources and feature sets. The goal of thispaper is to investigate the predictability of the personality traits of socialmedia users based on different features and measures. We extract social dataand questionnaire, and focus on how to use the user text information to predicttheir personality characteristics. We use the correlation analysis andprincipal component analysis to select the user information, and then use themultiple regression model, the gray prediction model and the multitasking modelto predict and analyze the results.
Keywords–Data Mining, PersonalityPrediction, Sentiment analysis, Social Media.