dump: Project Gutenberg... Roget2000

INTRODUCTION

Project Gutenberg in collaboration with MICRA, Inc released the 1911 version of Roget's Thesaurus as public domain electronic text (e-text).¹ [...] A problem with using the 1911 version of Roget's Thesaurus is that many modern technical terms are not in the thesaurus. The thesaurus also contains archaic terms that are not frequently used today. In addition, the category names were chosen by Roget and may not be obvious to a user.
Roget2000 also uses the e-text version of the Roget's Thesaurus. It is a Web based tool for semantic research that was implemented at Indiana University by Uta Priss and John Old.⁶ Roget2000 was proposed as a tool for semantic research by Dr. Rudolf Wille, founder of Formal Concept Analysis (FCA). This project follows conceptually from more than 30 years of research by Dr Walter A. Sedelow Jr. and Dr Sally Yeates Sedelow. The data used in this project is derived largely from a database version of Roget's Thesaurus built from Project Gutenburg's (sic!) free-text version of Roget's Thesaurus (version 13a - 1911 - copyright information), and WordNet (developed under the direction of Prof. George Miller).
In contrast to the previously mentioned search engines, Roget2000 is designed for graphical queries and graphical output, text queries and text output, and mixed queries. Currently, only text queries and outputs are available. For the text queries a user can choose from three methods: (1) searching for synonyms of a word, (2) searching for the semantic linking term(s) between two query words, and (3) browsing the hierarchy of concepts and corresponding antonyms. [...]

¹ This quote comes from the Gutenberg Philosophy http://www.promo.net/pg/history.html (Accessed 4/2001)
⁶ http://ella.lib.indiana.edu/~jold/Roget2000/index.htm (Accessed 4/2001)

Quelle: http://ella.slis.indiana.edu/~katy/paper/spie02-jason.pdf