A Comparative Analysis of Machine Learning Algorithms for Personality Traits Prediction

Abstract

Humans' personalities play a significant psychological role. One characteristic that affects how people interact with the outside environment is personality. A person's personality can be thought of as a crucial component of their behaviour. People's personalities are determined by how they connect with other people. Characteristic thoughts, feelings, and behaviour patterns are reflected in people's personality traits. This project aids in the creation of personality tests and the assessment of an individual's personality. The person can view their personality type and make improvements to it depending on the results of the personality classification. The goal of this study is to evaluate the performance of several classification algorithms for personality trait prediction. In this analysis, a system for predicting personality is built using the Big Five personality traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism. We provide a comparative analysis of the following data mining algorithms for personality trait prediction: decision tree, random forest, support vector machine, naive bayes algorithm, K-nearest neighbour, and artificial neural networks. We evaluate the performance of these algorithms on the Kaggle dataset. We compare all of the algorithms, the Random Forest algorithm has the highest accuracy (95.60%). Random forest outperformed the other five machine learning algorithms in terms of accuracy. Keywords: Decision Tree, Random forest, Support vector