Semantics-based threat structure mining

ABSTRACT

Today's National and Interstate border control agencies are flooded with alerts generated from various monitoring devices. There is an urgent need to uncover potential threats to effectively respond to an event. In this paper, we propose a Semantic Threat Mining approach, to discover threats using the spatio-temporal and semantic relationships among events and data. We represent the potentially dangerous collusion relationships with a Semantic Graph. Using domain-specific ontology of known dangerous relationships, we construct an Enhanced Semantic Graph (ESG) by scoring the edges of the semantic graph and prune it. We further analyze ESG using centrality, cliques and isomorphism to mine the threat patterns. We present a Semantic Threat Mining prototype system in the domain of known dangerous combination of chemicals used in explosives.

REFERENCES

• 1: S. Bechhofer, "The DIG Description Logic Interface: DIG/1.1", Available from http://dl-web.man.ac.uk/dig/2003/02/interface.pdf, 2003.
• 2: V. Haarslev, R. Moller, "RACER Users's Guide and Reference Manual", Available at http://www.cse.concordia.ca/%7Ehaarslev/racer/racer-manual-1-7-19.pdf, 2000.
• 3: V. P. Janeja, V. Atluri, J. S. Vaidya, and N. Adam. Collusion set detection through outlier discovery. In IEEE Intelligence and Security Informatics, 2005.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.4 Knowledge Representation Formalisms and Methods
          Subjects: Semantic networks

Additional Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Spatial databases and GIS

General Terms:
Design