Defender's Edge: Utilizing Intelligent Agent Technology to Anticipate Terrorist Acts

摘要

These briefing slides examines the need for, the potential of, and the steps involved in applying intelligent software, called Intelligent Agents (IA), to the problem of anticipating terrorist acts. Two areas in which IAs could play a significant role are in the development and maintenance of a real- time picture of terrorist activity, and in predicting potential occurrences of terrorist acts sufficiently well in advance so that they can be prevented. An examination of situation indicates the fragmented nature of the current flow of intelligence data from the collecting organizations, through the processing of the data into information, to the organizations designated to take action on it. Terrorism crosses operational and mission boundaries of the organizations established to counter it, such as the CIA, U.S. Customs Service, U.S. Immigration and Naturalization Service, FBI, and Department of Defense. Crime fighting organizations are reactive, crime investigation, and prosecution oriented, whereas terrorism requires a predictive, anticipatory, and preventative orientation. The multitude of data sources, coupled with the amount of data available to be collected can lead to information overload. These slides describe a way forward that includes both an approach to developing a data collection, data analysis, and information distribution plan, and a concept for utilizing intelligent agents to process the data collected into actionable information. The approach is illustrated by applying it to one possible means for terrorists to bring weapons into the United States, which is to hide them in a shipping container. About 17,000 shipping containers arrive in this country each day, only a small percentage of which can be searched. Thus, identifying the best containers to search for weapons or explosive materials is a complex task. This case study shows how IAs can be used to integrate and analyze data from key organizations to accomplish this task. (19 refs.).