Google is also close to natural
language understanding; it can answer simple questions up to a few words.

Not really.  Given a simple short question, it can present a human
with text that a human can recognize as containing the answer (among
other things).  That is pretty different from answering the question.

PowerSet, or some other scalable NL search method based on a fusion of
rule-based and statistical parsing, may end up being able to answer
simple short questions based on the massive knowledge on the Web
though.  I do think this is do-able with pre-AGI technology, but
Google isn't gonna get there with its current, non-NL-parsing-based
methodology.  (However, they have hired plenty of statistical NL
parsing experts, so they may well beat PowerSet there ... or acquire
PowerSet, etc.)

Ben G

-----
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=231415&user_secret=fabd7936

Reply via email to