Intermediate-Level Event Extraction for Temporal and Spatial Analysis

This task seeks to develop a deployable system for intermediate-level event extraction with emphasis on sense tagging and time/location normalization. Currently, only shallow Information Extraction (IE), mainly the identification of named entities andsimple events, is available for deployment. There is an acute demand for concept based, intermediate-level extraction of events and their associated time and location information. The results of this effort will be applied for information visualization tosatisfy the need for information analysts and decision makers. To extract concept-based, general events from free text requires the application of `cutting edge' Natural Language Processing (NLP) technology. The approach Cymfony proposes consists of ablend of machine learning techniques, hand-coded FST (finite state transducer) rules, and hard-coded procedures. By taking this flexible approach in combining different techniques and methods, it will be possible to exploit the best of different paradigmsdepending on the specific task being handled. The final deliverable includes (i) an IE system for intermediate-level extraction of concept based events, and (ii) an event visualization prototype to demonstrate the application of the new IE technology forinformation analysts.