Introduction to Information Extraction

The purpose of this site is to introduce Information Extraction and its usefulness to Internet users and information processing researchers. Information Extraction is the next step up from search engines in fulfilling information processing needs. For a definition of Information Extraction see Information Extraction. For an example of various forms of Information Extraction see Examples.

Information extraction systems have been evaluated under the support of DARPA and other government agencies for almost a decade. As the "Consumer Reports" for Information Extraction technology, these evaluations have set operational goals for the research, measured performance of implemented systems on real life tasks, and broadened the market for these systems. The evaluations were carried out beginning in 1987 by Beth Sundheim of the then Naval Ocean Systems Center (NOSC) and continued with DARPA funding under the TIPSTER Program for support by Dr. Nancy Chinchor of Science Applications International Corporation (SAIC). For the proceedings of the most recent evaluation of information extraction systems, see MUC-7 Proceedings.

This site also contains a variety of datasets and scoring software to benchmark systems independently. For the available benchmarking packages, see MUC Datasets.

To access the sites of major researchers in Information Extraction and related areas, see Major Research Sites.

For more website information contact: Ellen Voorhees
For more evaluation information contact: Nancy Chinchor
Last updated: Tuesday, 08-Mar-2005 13:14:36 MST
Date created: Friday, 12-Jan-01