On 11/10/07, Neil H. <[EMAIL PROTECTED]> wrote:

> The research is still quite early, but could Google Sets also be
> useful for more general AI tasks?

Only to the extent that simple first-order association by textual
proximity is useful, which is to say, only slightly.

Others have performed deeper experiments (search "mutual information"
google) but it's only superficially semantic, and a dead end in terms
of going any deeper.

- Jef

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=63900813-fe41d8

Reply via email to