Skip to Main Content
The quality of decisions made in business and government relates directly to the quality of the information used to formulate the decision. This information may be retrieved from an organization's knowledge base (Intranet) or from the World Wide Web. Intelligence services Intranet held information can be efficiently manipulated by technologies based upon either semantics such as ontologies, or statistics such as meaning-based computing. These technologies require complex processing of large amount of textual information. However, they cannot currently be effectively applied to Web-based search due to various obstacles, such as lack of semantic tagging. A new approach proposed in this paper supports Web-based search for intelligence information utilizing evidence-based natural language processing (NLP). This approach combines traditional NLP methods for filtering of Web-search results, Grounded Theory to test the completeness of the evidence, and Evidential Analysis to test the quality of gathered information. The enriched information derived from the Web-search will be transferred to the intelligence services knowledge base for handling by an effective Intranet search system thus increasing substantially the information for intelligence analysis. The paper will show that the quality of retrieved information is significantly enhanced by the discovery of previously unknown facts derived from known facts.