Stanford University

Using Automated Facial Expression Analysis for Emotion and Behavior Prediction

Ahn, S.J., Bailenson, J.N., Fox. J, & Jabon, M.E. (2010). Using Automated Facial Expression Analysis for Emotion and Behavior Prediction. in Doeveling, K., von Scheve, C., & Konjin, E. A. (Eds.), Handbook of Emotions and Mass Media (349-369). London/New York: Routledge.

View PDF

Abstract

In this chapter, we propose a model of approaching facial expression detection and analysis that goes beyond category-based measurements by incorporating automated technologies that make create and automatically improve prediction models based on raw facial feature movements. Using a computer equipped with a small camera, tracking software, and machine learning (an automated form of computer generated self-enhancement), we are able to select the most relevant facial features out of the massive collection of raw data and improve the prediction model in a time- and cost-efficient way. This allows us to utilize the raw data without fitting them into predefined categories, giving us greater analytical power than conventional category-based predictions. Moreover, our suggested methodology is notably unobtrusive when compared to other behavioral measures such as physiological measures, which usually require numerous sensors to be attached to the body.

Comments are closed.