Models of crowd behavior facilitate analysis and prediction of human group behavior, where people are affected by each other's presence. Unfortunately, existing models leave many open challenges. In particular, psychology models often offer only qualitative description, while computer science models are often simplistic, and are not reusable from one simulated phenomenon to the next. We propose a novel model of crowd behavior, based on Festinger's Social Comparison Theory (SCT). We propose a concrete algorithmic framework for SCT, and evaluate its implementation in several crowd behavior scenarios. Results from task measures and human judges evaluation shows that the SCT model produces improved results compared to base models from the literature. Copyright © 2007, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
|Proceedings of the National Conference on Artificial Intelligence
|Published - 28 Nov 2007