Personality trait prediction based on game character design using machine learning approach

摘要

This paper explores machine learning model to predict personality traits of game players based on game character design they choose before starting to play game. The data for this study is collected using survey technique. The total number of respondents for this study were 94 people, which all of them are game players with the minimum age of 15 years old. The experiment results show that our model can predict Openness, Conscientiousness, and Agreeableness personality traits from the game character design selected before playing games with the average accuracy of 66.7%.