To Buy or Not to Buy? Understanding the Role of Personality Traits in Predicting Consumer Behaviors.

In this paper, we explore the role of derived personality traits from Twitter in determining consumer behaviors. We conduct comprehensive analysis on a large industry dataset containing 188,654 individual level purchasing records across over 100 product categories. For each category, we build classifiers to distinguish buyers from non-buyers based on their derived personality traits. We use the models with demographic features as baseline to evaluate the predictive power of personality traits. Our results prove the decisive power of derived personality on a variety of consumer behaviors.