An ontology represents knowledge about a particular domain. This knowledge is a description of the type of entities, relationships, terms and concepts that constitute knowledge about a particular domain. Such a domain ontology reflects the areas of concern of the analysts, the kind of role they have within their organization, the specifics about their goals and the tasks they perform.

The ontology component represents a critical piece in solving the analyst’s information overload problem. By leveraging an understanding of the user’s context – the essence of the entities, concepts, relations, tasks and goals of the user – an ontology maximizes the effective application of the analyst’s attention.

Ontologies and RiverGlass

Ontologies are used in multiple ways within RiverGlass products. One example is how relevance is computed in the open source intelligence acquisition from the web. RiverGlass software traverses large parts of the open web, but only collects information when it is deemed relevant to an analyst’s needs. The ontologies play a primary role: if a piece of web content has many of the terms or concepts that exist in part of an ontology, then that content is deemed more relevant to an analyst’s needs than a page which has few or none.

Ontologies are also used as one part of automated information extraction, the process in which entities (people, places, things, groups, etc.), relationships and events are recognized and identified within any unstructured text. Information extraction is used directly by analysts, and is also input to other analytical procedures within the system.