Analysing the Personality traits using machine learning techniques
Murari Devakannan Kamalesh, B. Bharathi · 2022 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES) · 2022
The term "personality" refers to a person's unique set of traits that influence their habits, behaviours, attitudes, and mental patterns. As part of this research, XGBoost is used to predict four personality traits based on the Myers-Briggs Type Indicator (MBTI) model[1] , Feeling from input text using Machine Learning Techniques, namely XGBoost classifier XGBoost. The experiments make use of a publicly available benchmark dataset from Kaggle. Prior study has revealed that the dataset has a high degree of skewness. This can be reduced using the Re-sampling technique, which employs random oversampling. Also used for character analysis is text pre-processing techniques such as tokenization (word stemming), stop-word deletion (stop words), and feature selection. This research lays the groundwork for the creation of a system for identifying people's personalities, which might be used by organisations to better attract and select employees and to better serve their clients. Overall, the results of all classifiers across all personality qualities are acceptable, but the performance of the XGBoost classifier stands out by obtaining precision and accuracy of above 99 percent for various personality features.