Personality Traits on Twitter—or—How to Get 1,500 Personality Tests in a Week

Psychology research suggests that certain personality traits correlate with linguistic features. This correlation can be effec-tively modeled with statistical natural lan-guage processing techniques. Prediction accuracy of these models should improve with larger data samples and more fea-tures. Most existing work on personality prediction from text, however, focuses on small samples and closed-vocabulary in-vestigations. Both factors limit general-ity and statistical power of the results. In this paper, we explore the use of social media as a resource for large-scale, open-vocabulary personality detection. We ana-lyze which features are predictive of which personality traits, and present a novel cor-pus of 1.2M tweets with personality and gender annotation. Our results suggest that social media can be a valuable source for certain personality type predictions. 1