Personality detection from text, based on the MBTI model

Abstract

Personality is a person's distinguishing set of behaviours, ways of perception and emotional patters. It also plays a key role in everyday life, and the addition of personality awareness across various fields may be of great benefit. The idea of obtaining a person's personality type without having to go through lengthy and at times biased traditional methods of questionnaires and interviews is thus of interest. With the growing popularity of online social networking sites, it is no longer difficult to get a hold of text generated by users of the various platforms. And with the advances in Artificial Intelligence (AI), it is now possible to make use of machine learning algorithms to detect personality. In this project, personality detection based on the Myers–Briggs Type Indicator (MBTI) personality model is explored using various machine learning algorithms. Data is first pre-processed and prepared to train the various machine learning algorithms that will form the classification models. The performance of each model is then recorded by testing them against data that has not been used for training the models. The model that performed the best can thus be evaluated and improvements can be made upon the model to increase accuracy in future work.