Modeling current events

Enormous amounts of information are produced every day, all over the world, but very little of it is true. In this paper, we describe the modeling component of a current events analysis system that combines statistical and Knowledge-Based approaches to natural language analysis with new modeling and architecting techniques to produce a Knowledge-Structured Virtual Subject World. This Multi-User Virtual Environment (MUVE) allows groups of subject- or area-experts to collect and analyze relevant information, for visualization, exploration, consistency checking, and discovery. It supports the analyst's expertise and experience by emphasizing relevant information and filtering out irrelevant information. The system makes essential use of Computational Reflection, which is the ability of a system to analyze and adjust its own behavior. We briefly describe the architecture of the system, show how reflection leads to flexibility of operation, and concentrate on the modeling techniques that we use to collect, retain, and examine information.