Processor for Open source INTelligence (POINT)

Knexient proposes POINT2, a processing system for OSI on the Web. Specifically, POINT2’s text processing engine finds primary types—main events or entities discussed in online news. When the primary type is an event, it is related to its location and time of occurrence. The location of the primary event is converted into lat/lon coordinates and presented on KnexientGlobe, a custom geospatial user interface. This enables POINT2 users to quickly aggregate relevant information about events and their locations. POINT2 uses an innovative hierarchical classification system, the Document Classification System (DCS), to classify primary events down to the granularity of particular actions (e.g., a suicide bombing), and to classify very large volumes of semantically heterogeneous text, such as reports about meetings, bombings, elections, or vehicle accidents. Specifically, DCS classifies news according to the primary specific event (e.g., a suicide bombing), and the event context (e.g., the Iraq War). The event pair tells the POINT2 user what the story is about based on the events described. Non-events such as primary persons or organizations mentioned in online news are also identified by DCS.