5.3
JPT Research Digest
Machine Learning Approach to Personality Type Prediction.
Amirhosseini, M. H. and Kazemian, H. (2020). Machine learning approach to personality type prediction based on the Myers-Briggs Type Indicator®. Multimodal Technologies and Interactions, 4(9). https://doi.org/10.3390/mti4010009 |
This study developed new machine learning methods for automating personality prediction with a goal to assist Neuro Linguistic Programming (NLP) practitioners and psychologists. NLP develops personality in individuals by working with metaprogrammes: implicit cognitive strategies people use to sort information and make decisions leading to specific behaviors. This study aims to create a new algorithm to predict personality using a Myers-Briggs personality dataset from Kaggle, an online environment where users find and publish large datasets. The dataset has two columns: one indicates the Myers-Briggs type of the individual and the other contains 50 posts on social media from that same user.
The authors' algorithm produced accurate prediction rates of 67% to 86% for individual type preferences–a similar achievement to a linguistic analysis of a MBTI® Reddit community conducted by Gjurkorić, M. and Šnajder, J. (2018). In that study (highlighted in the 2019 JPT-RD) personality prediction was accurate 67% to 82% for individual preferences and 82% accuracy was achieved for three or more preferences of whole type.
Both personality prediction studies relied on datasets where the users already knew their MBTI type. Valuable future research would predict personality type from datasets of people who did not know their type when the dataset was produced.
ARTICLE PERMALINK: https://www.myersbriggs.org/research-and-library/journal-psychological-type/machine-learning-approach-to-personality-type-prediction/
ARTICLE COMMENTS: