The real-time nature and value of homeland security information

ABSTRACT

Ensuring the security of our homeland depends in large measure on two quite distinct factors: having the knowledge necessary to prevent, predict, prepare for, or respond, if necessary, to any manner of terrorist attack or a natural or manmade disaster and collaborating or sharing knowledge with a broad range of international, federal, state, local, and tribal agencies, as well as other private or public organizations. The essential problem with adequately addressing these factors, despite the many advancements made in the past decade, is twofold. First, it is not so much the mass but rather the diffuse nature and complexity of the data, information, and knowledge required for understanding terrorism and accounting for the manifold consequences of disasters that make possession of the right knowledge difficult. And, second, that diffuseness and complexity is magnified by the extreme diversity and wide distribution of the many potential homeland security collaborators. Retrospective analysis, and even knowledge discovery, is less useful under these conditions than prospective, real-time synthesis of information for multiple users. Also, privacy is as important as, if not more important than, security. This suggests that database designs and techniques for information retrieval, and knowledge management must take advantage of such technologies as semantic nets, visualization, and discrete mathematics to build knowledge systems capable for homeland security applications.

Index Terms

• Computing methodologies

  • Artificial intelligence

    • Knowledge representation and reasoning

      • Semantic networks

• Information systems

  • Information retrieval

    • Retrieval tasks and goals

      • Document filtering

      • Information extraction

  • Information systems applications

    • Data mining