A graph-based framework for fusion: From hypothesis generation to forensics

Moises Sudit, Adam Stotz, Rakesh Nagi, Kedar Sambhoos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The intent of this paper is to show enhancements in Level 2 and 3 fusion capabilities through a new class of graph models and solution strategies. The problem today is not often lack of information, but instead, information overload. Graphs have demonstrated to be a useful framework to represent and analyze large amounts of information. Classical strategies such as Bayesian Networks, Semantic Networks and Graph Matching are some examples of the power of graphs. We will introduce two different but related graph-based structures that will allow us to span the temporal performance of decision-making processes. Given that most of the high level information fusion problems of interest are NP-Hard, there is a need to separate methodologies between "near real-time" tools and forensic heuristics. With this in mind we will introduce a real-time decision-making tool (INFERD) and a forensic graph matching algorithm (TruST).

Original languageEnglish (US)
Title of host publication2006 9th International Conference on Information Fusion, FUSION
DOIs
StatePublished - Dec 1 2006
Externally publishedYes
Event2006 9th International Conference on Information Fusion, FUSION - Florence, Italy
Duration: Jul 10 2006Jul 13 2006

Publication series

Name2006 9th International Conference on Information Fusion, FUSION

Other

Other2006 9th International Conference on Information Fusion, FUSION
CountryItaly
CityFlorence
Period7/10/067/13/06

Keywords

  • Data graph
  • Graph matching
  • Hypothesis
  • INFERD
  • Situational awareness
  • Template
  • TruST

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems

Fingerprint Dive into the research topics of 'A graph-based framework for fusion: From hypothesis generation to forensics'. Together they form a unique fingerprint.

Cite this