Computer assisted security assessment through fact proposition space

Peter Hospodka, Qiuming Zhu, William Sousan, Ryan Nickell

Research output: Contribution to conferencePaper

Abstract

In this age of Information Warfare, it is becoming increasingly important to make quick and informed decisions based on current information and knowledge gained from the past. Whether these decisions are used for assessing risk involved with an action, threat involved with an opponent's action, or impact of your decision on opponent's actions, it is a daunting process without support from advanced situational assessment systems. To provide such support and assistance, we have devoted our effort on an intelligent inference system operating based on a hierarchical fact-proposition space (FPS) model. Basic components of our FPS consist of a set of propositions or questions that need confirmation or answers at the levels of the hierarchy. Some propositions have relationships where the outcome of one affects the outcome of another. The FPS inference process tries to use the past knowledge to determine initial beliefs or strength of the relationships. Current knowledge is further used to identify some proposition confirmation or question answers which in turn are used to propagate up the hierarchy in the goal to return some recommended decision(s). Our FPS model has similar foundations of Bayesian Networks (BN) in that both base computations on Bayes' Theory and probabilistic inference. However, our FPS differs from BN in that it sub-divides a complicated multi-level, multi-variant decision problem into a number of hierarchically organized subspaces, where each space can be manipulated in terms of matrix computations in a simple form of Bayesian inference. By partitioning the solutions at the subspaces in the hierarchy, the total complexity of the problem is reduced. The system's objectives are to provide a valuable information fusion and belief integration engine to those making decisions on risk assessment, threat assessment, and similar tasks. Through the prototype implementations, we hope to demonstrate the capabilities and advantages of the hierarchical fact-proposition space model.

Original languageEnglish (US)
Pages195-202
Number of pages8
Publication statusPublished - Jan 1 2008
Event3rd International Conference on Information Warfare and Security, ICIW 2008 - Omaha, NE, United States
Duration: Apr 24 2008Apr 25 2008

Conference

Conference3rd International Conference on Information Warfare and Security, ICIW 2008
CountryUnited States
CityOmaha, NE
Period4/24/084/25/08

    Fingerprint

Keywords

  • Computer assisted assessment
  • Decision making
  • Fact proposition space
  • Inference engine
  • Information warfare

ASJC Scopus subject areas

  • Information Systems
  • Safety, Risk, Reliability and Quality

Cite this

Hospodka, P., Zhu, Q., Sousan, W., & Nickell, R. (2008). Computer assisted security assessment through fact proposition space. 195-202. Paper presented at 3rd International Conference on Information Warfare and Security, ICIW 2008, Omaha, NE, United States.