- From: Danny Ayers <Danny.Ayers@highpeak.ac.uk>
- Date: Mon, 06 Dec 1999 15:06:04 +0000
- To: wordnet@princeton.edu, "www-rdf-interest@w3.org" <www-rdf-interest@w3.org>, danbri@w3.org
Hi, A long term project I'm working on (& off!) is an intelligent web page classifier/search engine. While gathering resources, two of the potentially most significant I found are the Open Directory (http://dmoz.org/) and WordNet (http://cogsci.princeton.edu/~wn/). I had the idea (not original!) of accessing/storing the data in a form of conceptual space (or semantic web, if you like), using Kohonen SOMs/ LVQs, so that pages with similar concepts would be linked/close in indexing terms. I was thinking in terms of using RDF in the back end, probably tied to a SQL database for the site indexing. At the front (or middle!?) I was thinking in terms of using WordNet as an extended thesaurus for 1. more natural language understanding of search queries, 2. clearer 'conceptual' indexing of the web pages. It had never occurred to me to restate the WordNet data in RDF - in a system described as above, this would make for wonderful consistency & versatility. Has anyone any thoughts on how the word(/concept) space & web space cataloguing could be integrated? Cheers, Danny. ------------- Click on a button on this site to make a *free* food donation to hungry people (paid for by sponsors) : http://www.thehungersite.com/index.html
Received on Monday, 6 December 1999 10:07:57 UTC