An Evaluation of Machine Learning Techniques for Personality Classification Using the Myers–Briggs Assessment

Personality classification is an emotional and psychological perception that involves the classification of individuals on the basis of their behavioral, emotional, and cognitive patterns. Myers–Briggs type indicator (MBTI) is one of the popular personality classification frameworks, which comprises sixteen categories of personalities into four classes such as extraversion/introversion, sensing/intuition, thinking/feeling, and judging/perceiving. The study of personality assessment can be useful for individuals as well as for the different organizations. It can provide valuable insights in various domains like academics, hiring employees, and predicting criminal mind analogy. In this paper, various machine learning techniques, namely support vector machine, XGBoost, random forest, and logistic regression, have been employed. The results revealed that XGBoost performed well as compared to other classifiers with the 67% accuracy for collective data which includes posts and MBTI types as classes and support vector machine is better with the higher accuracy of 77.96, 86.03, 72.62, and 65.87 for IE, NS, FT, and JP categories, respectively.