- From: Sampo Syreeni <decoy@iki.fi>
- Date: Fri, 1 Aug 2003 11:45:27 +0300 (EEST)
- To: "Judson, Ross" <rjudson@managedobjects.com>
- cc: www-rdf-interest@w3.org
On 2003-07-31, Judson, Ross uttered: >Now imagine that someone builds a big corpus of URIs, and starts to >categorize them. Someone like, say, Google, or DMOZ. Interesting. Suppose you use the pages in ODP categories to train your classifier (Bayesian or not), then run it over Google's cache. That'd help Google directly, because topic clusterings ought to make good related pages listings after being prioritized by PageRank and what else Google now uses. They just might run with something like that, especially since properly constructed Bayesian classifiers can be trained in linear time, and incrementally. The results could be exposed as triples grafted onto the ODP hierarchy. -- Sampo Syreeni, aka decoy - mailto:decoy@iki.fi, tel:+358-50-5756111 student/math+cs/helsinki university, http://www.iki.fi/~decoy/front openpgp: 050985C2/025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2
Received on Friday, 1 August 2003 04:45:31 UTC