For better understanding visit this blog: BLOG
In this project we have used the Myers-Briggs Personality Type (MBTI)Indictor, which describes a person’s personality types designed by Isabel Myer and Katherine Briggs based on Carl’s Jung theory of personality types. It consists of 16 unique kinds of personalities combination of 4 dimensions which are Extraversion (E) or Introversion (I), Sensing (S) or Intuition (N), Thinking (T) or Feeling (F), and Judging (J) or Perceiving (P) which also defines one’s strengths and preferences.
We have implemented Natural Language Processing techniques, Various Machine Learning and Deep learning models to precisely predict one’s MBTI personality type indicated by the four-letter code.
We have consumed the Kaggle dataset, which is publicly available contains in the form of a CSV file with 8675 rows and two columns(features) which are posts that have 50 social media posts gathered with ‘|||’ between them and other is the type which is one of the 16 MBTI types of corresponding users.
We used models such as: SVM, RandomForest, XGBoost, CatBoost, and KNN and computed their F1-Scores.