“Cognitive Factors in Homeland Defense” Reusing Intelligence Analysts' Search Plans

Abstract

Because of the huge volumes of data that open source intelligence analysts must search, effective information gathering on the web, is a complex activity requiring planning, text processing, and interpretation of extracted data to find information relevant to a major intelligence task or subtask (Knoblock, 1995), (Lesser, 1998) and (Nodine, Fowler et al., 2000). This paper describes our use of analysts' plans to build a case library of reusable automated plans. This process involves analyzing analysts' approaches to identify the important characteristics of a knowledge discovery plan for case indexing and retrieval and devising techniques for representing plans so that they can be automatically adapted and executed. For this research, we are not building a “complete” case library, but are developing approaches and guidance for getting maximum usage out of the cases that are represented and providing processes for identifying further areas for case development. The knowledge discovery plan categories that we have identified thus far correspond to different contextual domains that may be indicators of terrorist activities. When a plan is selected from one of these categories, it is adapted, and the resulting plan is executed, relevant information is extracted from unstructured documents, and the extracted information is used to make further inferences and launch additional searches.