Using a text analysis and categorization tool to generate Bayesian belief networks for use in cognitive social simulation from a document corpus
Daniel C. McKaughan, Zachary Heath, Jonathan T. McClain · Annual Simulation Symposium · 2011
Cognitive social simulation provides a promising means of gaining insight into a particular area of interest for a given population, but the data to instantiate these models must be gleaned from disparate data sources. Methods and tools to efficiently leverage information regarding a population of interest from a document corpus are required. This paper provides an overview of the use of Sandia National Laboratory's Text Analysis Extensible Library (STANLEY) to categorize a body of documents to create Bayesian belief networks that can be used as cognitive models within social simulation.