Challenges of Country Modeling with Databases, Newsfeeds, and Expert Surveys

Abstract

According to expert practitioners and researchers in the field of human behavior modeling, a common central challenge now confronting designers of HBM (human-behavior-modeling) applications is to increase the realism of the synthetic agents' behavior and coping abilities. In this chapter, we have argued that the available country datasets are an invaluable resource that will permit us in the human behavior M&S (Modeling and Simulation) field to more realistically profile factions, and their leaders and followers. This in turn will help us to develop tools for those interested in analyzing alternative competing hypotheses for DIMEPMESII (Diplomatic, Informational, Military, and Economic actions - Political, Military, Economic, Social, Informational, and Infrastructure effects) studies. At the same time, there are significant growing pains and challenges involved in trying to put the country data to use. This chapter reviewed those challenges by looking at three pathways for extracting and parameterizing the data - webscraping of newsfeeds, extracting and translating data from country databases, and (semi-) automated surveying (i.e., web questionnaires with data translation and model instantiation capacity) of subject matter experts. In each of these areas there are significant challenges and obstacles to seamless integration, not the least of which is that profiling individuals and groups is difficult even for the smartest humans. By using a triangulation of the three approaches, and a knowledge engineering approach that mimics how country and leader experts currently do the job (alternative competing hypotheses), we believe that one can move ahead as outlined in this chapter. This chapter examined how to use this approach with the help of a case study involving a socio-cognitive agent architecture (FactionSim-PMFserv).

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2008
Accession Number
ADA519799

Entities

People

  • Barry G. Silverman
  • G .j. Kim
  • Guana K. Bharathy

Organizations

  • University of Pennsylvania

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Agent-Based Simulations
  • Cognition
  • Cognitive Science
  • Cognitive Systems Engineering
  • Data Sets
  • Databases
  • Digital Information
  • Employment
  • Human Behavior
  • International Organizations
  • Minority Groups
  • Personnel Management
  • Political Science
  • Psychology
  • Recreation
  • Social Sciences
  • Terrorists

Readers

  • Computational Modeling and Simulation
  • Educational Psychology
  • Systems Analysis and Design