Newsgroups: comp.theory.info-retrieval
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!usenet.eel.ufl.edu!news.ultranet.com!zombie.ncsc.mil!simtel!harbinger.cc.monash.edu.au!bunyip.cc.uq.oz.au!munnari.oz.au!news.hawaii.edu!news
From: Anne Veling <anne@mrnews.mro.dec.com>
Subject: Re: Meta-word clustering methods?
X-Nntp-Posting-Host: uhunix3.uhcc.hawaii.edu
Message-ID: <3pher8$q5a@mrnews.mro.dec.com>
Lines: 23
Sender: news@news.hawaii.edu
Organization: Digital Equipment Corporation - Marlboro, MA
References: <3o80om$m58@sofya.math.byu.edu>
Date: Fri, 19 May 1995 06:46:00 GMT
Approved: Pollarda@uhunix.uhcc.hawaii.edu


This is what I mean by meta-words: words that are in some way semantically
related; words that appear in the same documents, in the same sentences or
contexts. In this way, meta-words are some sort of high-level clusters of
related words: synonyms e.g.

The question still stands: do you know of any work done in this area?

Thanks,

Anne.
-- 
############################################################
W.A. Veling (Anne)		veling_a@cecamo.enet.dec.com
Nieuwluststraat 46		veling@cs.utwente.nl
7523 XB Enschede
The Netherlands			Interested in: AI, philosophy, reading,
Computer Science Student	cooking, windsurfing, whisky, bible,
University of Twente		interactive fiction, ethics, anything
############################################################
What you get out of it, must have been in it.
(applicable to life, art & books)
############################################################

