Performance Comparison of Machine Learning Algorithms for Student Personality Classification

Everyone has their own characteristics and personality. The questionnaire instrument used to measure a person's personality was developed by Costa and McCrae in 1992, known as the Big-Five Personality model. This instrument consists of 50 statement items using a 5-point Likert scale rating. The purpose of this study is to analyze the performance of each Machine Learning algorithm such as Support Vector Machine (SVM), Random Forest (RF), and Neural Network (NN) for student personality classification based on the OCEAN big five personality models consisting of Openness (O), Conscientiousness (C), Extraversion (E), Agreeableness (A), and Emotional Stability or Neuroticism (N). The results showed that the Neural Network method was able to produce the best accuracy value of 76% and was followed by the Random Forest and SVM methods with an accuracy value of 56% and 40%. Recognizing the personality of oneself and others can determine the pattern of interactions and reactions carried out, including patterns of interaction in learning activities between teachers and students. Furthermore, it can be investigated the ability of machine learning algorithms to predict student academic performance based on their character and personality.