> Ben Goertzel wrote:
>
> >...
> >
> >As I said in a recent post, I prefer to stay away from natural language
> >processing at this stage, until the system has acquired a rudimentary
> >understanding of natural language thru its own experience.
> We're not quite
> >there yet ;)
> >
> >ben
> >...
> >
> I wasn't thinking of a NLP system, but rather of a system that could
> classify e-mails by variety, and could assemble possible replies out of
> boilerplate appropriate to the key terms.  Full understanding isn't
> needed by this kind of system, but it could still be quite useful.  And
> expand easily to cover a larger domain, as techniques advanced.

OK, I understand.  There are actually many products out there that do this,
with moderate levels of success.

One of the nicest was made by a company called Artificial Life, which died 8
months ago or so.  They also made chat tools aimed at the online customer
support market.

I don't remember the names of the other firms making similar products, but
there are lotsa them.  This is part of the eCRM market sector (CRM =
Customer Relationship Management).

I still don't think this is a great app for a proto-AGI because

-- moderate performance is obtainable with narrow-AI methods

-- AGI can really contribute something beyond narrow-AI, here, only once it
has a deep understanding of human language

But of course that's just my intuition, and I could be wrong....  This
intuition is perhaps more pertinent to Novamente than to other proto-AGI
systems.

-- Ben G

-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/

Reply via email to