One part of this effort involves the creation of a standard architecture (standard interfaces) for systems doing information retrieval and extraction. This architecture is being developed by a group of contractors and Government personnel (the Tipster Architecture Working Group), chaired by Ralph Grishman of NYU.
The current version of this architecuture (Version 1.50) was distributed at the 12-month Tipster meeting, May 17-19, 1995. Three files are available:
Several additional documents have been prepared by the Tipster Architecture Committee (which consists of representatives of the Government agencies involved in the Tipster program) to define the goals of the architecture effort and the process for qualifying systems as "Tipster compliant":
(For information about other natural language processing research at the NYU Proteus Project, click here).