Related Experiment Video
Updated: Aug 6, 2025

Brain Imaging Investigation of the Neural Correlates of Observing Virtual Social Interactions
Published on: July 6, 2011
How social media expression can reveal personality
Nuo Han1,2,3, Sijia Li4, Feng Huang1
1Chinese Academy Sciences Key Laboratory of Behavioral Science, Institute of Psychology, Chinese Academy of Sciences, Beijing, China.
This study enhances machine learning personality assessment by integrating domain knowledge, improving model interpretability and accuracy. Findings link personality traits to mental health, offering new psychiatric applications.
Area of Science:
- Psychology
- Computer Science
- Artificial Intelligence
Background:
- Personality psychology investigates individual differences, with recent machine learning advancements focusing on online personality assessment.
- Interpretability of machine learning models for personality prediction remains a challenge, limiting understanding of assessed personality aspects.
- This study addresses the need for interpretable personality prediction models by incorporating domain knowledge.
Purpose of the Study:
- To develop and validate a machine learning model for personality assessment that enhances accuracy and interpretability.
- To investigate the role of domain knowledge in improving personality prediction models.
- To explore the relationship between personality traits and mental health through interpretable machine learning.
Main Methods:
- Recruited participants via an online platform and collected Weibo posts.
- Extracted textual features using six psycholinguistic and mental health-related lexicons.
- Developed a personality prediction model using the multi-objective extra trees method with 3,411 data pairs, evaluating validity, reliability, and feature importance.
Main Results:
- Features from the Culture Value Dictionary were identified as the most significant predictors.
- Fivefold cross-validation for personality trait prediction yielded scores between 0.44 and 0.48 (p < 0.001).
- Correlation coefficients for five personality traits between split-half datasets ranged from 0.84 to 0.88 (p < 0.001), indicating good model performance and contractual validity.
Conclusions:
- Integrating domain knowledge into machine learning models enhances both the reliability and interpretability of personality assessment.
- The study elucidates personality aspects measured by prediction models and establishes a connection between personality and mental health.
- This research highlights the potential of combining machine learning with domain knowledge in psychiatry and mental health applications.
Related Concept Videos
Cultural Influences on Personality
Emotional Expression
Universal Facial Expressions
Psychologist Paul Ekman identified seven basic...
The Sense of Self: Reflected Self-Appraisal and Social Comparison
Therapeutic Communication
Verbal communication depends on language or a prescribed way of using words so that people can share information effectively. The critical aspects of verbal...
Self-Presentation: Self-Monitoring and Self-Handicapping
Social Cognitive Perspective on Personality

