A Dialectic Approach to Intelligence Data Fusion For Threat Identification

Period of Performance: 08/13/1997 - 08/13/1999

$712K

Phase 2 SBIR

Recipient Firm

Bevilacqua Research Corp.
4901 Corporate Drive Array
Huntsville, AL 35805
Principal Investigator

Research Topics

Abstract

The successful Phase I demonstration showed that the new Bounded Neural Network (BNN) architecture could accurately and efficiently simulate human complex decision making skills. The application demonstrated in Phase I was aimed at automating human decision making in a selected foreign threat simulator. Instead of expanding on a single application of the BNN for the Government under Phase II, the BRC team's technical approach concentrates on creating the tools necessary to build and apply the BNN to a wide variety of systems. This will allow the Government to utilize the BNN architecture across the DoD. The overall objective of this Phase II program will be to create a militarized toolkit that contains easy-to-use tools that can be used to build intelligent BNN-based software applications for information handling. In addition, the team will look toward Phase III in this development process to facilitate the development of a full-scale, marketable, commercialized BNN toolkit. The toolkit will contain tools to build BNN's and an innovative BNN-based context level search engine for military and commercial Internist. Raving these tools at their disposal, Government analysts will be able to easily build and apply BNN-based intelligent processing software to a wide variety of different programs and products. There is a tremendous need on the World Wide Web for a context-based search engine to replace the existing text-based products. The BRC product would replace products like YAHOO, ALTA-VISTA, webcrawler and others currently on the web.