Predicting MBTI personality of YouTube users

Area of Science:

• Computational Social Science
• Psychological Informatics
• Natural Language Processing

Background:

• Personality assessment traditionally relies on self-report questionnaires.
• Emerging computational methods offer alternative approaches to personality analysis.

Purpose of the Study:

• To develop and apply a machine learning classifier for predicting Myers-Briggs Type Indicator (MBTI) personality types.
• To analyze the personality types of YouTube users commenting on specific content genres.

Main Methods:

• Utilized Natural Language Processing (NLP) and Machine Learning (ML) to build a personality classifier.
• Extracted text samples from YouTube comments for analysis.
• Classified users into MBTI types based on comment content.

Main Results:

• Accurate personality type estimation is achievable with approximately 100 topic-relevant words per user.
• INFP emerged as the most prevalent MBTI type among analyzed YouTubers.
• INTP was the predominant type for conspiracy theory comments, while INFP dominated spiritual and travel content discussions.

Conclusions:

• Computational analysis of social media text provides a viable method for personality type assessment.
• Introverted and Intuitive (N) personality traits are significantly represented in online commentary.
• Content topic influences the distribution of personality types observed in online user engagement.