ANALYSIS OF SOCIAL MEDIA IMAGES TO PREDICT USER PERSONALITY ASSESSMENT

Social networking usage is one of the most popular online activity today. The social media platforms are a great way and get to know and connect people. With the advent of internet and increasing use of internet technology huge amount of data are being uploaded to the internet today. Through the recent advancement in technology and access of mobile phones and internet with images have become widespread over social media. Users pick their choice of profile pictures; they post their daily routines on the social media platforms through text or images. Their choice of profile pictures and the type of image they post can be highly related to their personality. The purpose of this study is to develop a system which can predict a person's personality based on social media user details. The Big Five Model is the personality model used in this research. This method that integrates image features of user profile picture and analyze the personality traits of individual user. Currently, the lack of existing system to analyze and monitor the user's personality based on image is not being addressed. The goal of this study is to identify the way and extent to which differences in personality, measured using the Big Five model, are related to the choice of profile images. This provides a consistent reduction of a prediction error with respect to personality traits for each model. The results succeeded to outperform the models with the root mean square error ratio of 0.017.