Information DiscovEry Assistant that Learns (IDEAL)
Navy SBIR FY2011.2


Sol No.: Navy SBIR FY2011.2
Topic No.: N112-152
Topic Title: Information DiscovEry Assistant that Learns (IDEAL)
Proposal No.: N112-152-1304
Firm: Knowledge Based Systems, Inc.
1408 University Drive East
College Station, Texas 77840-2335
Contact: Perakath Benjamin
Phone: (979) 260-5274
Web Site: www.kbsi.com
Abstract: Traditional keyword search fails to adequately meet the needs of the modern intelligence analyst. If analysts were able to express their information needs in plain terms, understood by a search engine as guidance or examples, documents and other information artifacts might be brought to light that simply "guessing at" appropriate keywords would never elicit. Further, because analysts seldom work in isolation, a shared understanding of analysis goals and subsequent sharing of knowledge and effort can significantly improve analytic outcomes. KBSI proposes to design, configure, and demonstrate an innovative solution called the Information DiscovEry Assistant that Learns (IDEAL) that addresses the challenges described above. IDEAL will be designed to provide the following advanced capabilities (1) perform semantic search across multiple data sources to discover data that is relevant to an analyst's goals and tasks; (2) progressively refine the quality of information produced through iterative semantic search and knowledge discovery; (3) facilitate dynamic information sharing and collaboration by a group of human agents; and (4) provide learning and adaptation by using the results and analyst feedback from the previous search/analysis activities to incrementally refine the semantic mapping and search models over extended time.
Benefits: The primary short-term benefits include (1) Immediate increase in both precision and recall of searches performed by intelligence analysts, resulting in a significant reduction in the time required to discovered relevant information, (2) Significant increase in information sharing among analysts, reducing redundant search efforts and increasing the overall quality of information discovered by all collaborating analysts, and (3) Significant increases in precision and recall of analysts searches over time as the IDEAL system better learns the search tasks of a particular user through both explicit and implicit user feedback.

Return