A Fragment-Based Approach to Object Representation and Classification
IWVF-4 Proceedings of the 4th International Workshop on Visual Form
Artificial Intelligence: A Modern Approach
Artificial Intelligence: A Modern Approach
An introduction to variable and feature selection
The Journal of Machine Learning Research
Network Science: Theory and Applications
Network Science: Theory and Applications
Social and Economic Networks
A characterization of interventional distributions in semi-Markovian causal models
AAAI'06 proceedings of the 21st national conference on Artificial intelligence - Volume 2
Social Computing, Behavioral Modeling, and Prediction
Social Computing, Behavioral Modeling, and Prediction
Proceedings of the 2011 Military Modeling & Simulation Symposium
Representing dynamic social networks in discrete event social simulation
Proceedings of the Winter Simulation Conference
Hi-index | 0.00 |
Authentically representing large social collectivities remains a preeminent challenge throughout the social computing, and modeling and simulation communities. We demonstrate here a simple technique that uses survey and polling data to embed agents with attributes and endogenously elicit an authentic and theory-driven simulation social structure for an artificial society. We furthermore show that a representation of social structure based on internal agent attributes allows for the continuous representation of social dynamics that affect agent cognition and association, and that social structures for artificial societies can be generated without any loss to the granularity of the underlying data or simulation output. We provide a case study using social survey data to demonstrate the method and effects, document the visualization of social structure for the population of Indonesia, discuss the implications and uses of survey data for social simulation, and suggest several paths forward for social and behavioral predictive modeling.