Building Open-domain Semantic Search (BOSS)

The quantity of text available to Navy Intelligence analysts has become too large for any single person to read, identify all relevant information, and fuse the discovered information with the existing knowledge base. Much of the information is text. The challenge of exploiting high volume text is especially significant for military, government and private sector analysts. To actually perform analysis within narrow operational deadlines, analysts first need to identify relevant information against a background of high levels of noise. In the Building Open-domain Semantic Search (BOSS) project, we are creating a next-generation semantic search capability that efficiently indexes all concepts expressed in text. By using raw data to develop a model of possible concepts, BOSS can easily adapt to many domains and languages. With indexed concepts, users can search for ideas they are interested in, rather than being restricted to keywords, which can be ambiguous and incomplete. In addition to indexing concepts, BOSS also indexes structured data, including entities, events and relations between them. The goal of the Phase II effort for the Navy is to refine technologies developed in prior efforts for the Air Force and place them in the hands of operational users as rapidly as possible.