On 4/19/07, Benjamin Goertzel <[EMAIL PROTECTED]> wrote:
> The whole problem of reference resolution (eg pronouns) can be handled
by
> rules. Also, preposition resolution can be handled by rules, with
domain
> knowledge specified within the rules.
>
> Comments?
My comment is that hundreds of smart PhDs have been banging away on
precisely this approach for many years, and publishing papers on it;
and it seems clear to nearly all of them by now that this sort of
approach can only get you sooo far and then runs out of steam.
I don't yet see what new idea you are bringing to the approach...
There seems to be 2 current alternatives: statistical parsers and rule-based
NL systems.
Rule-based systems are brittle, time-consuming to engineer, and have narrow
coverage of sentences. Statistical parsers can be faster to build, have
broad coverage, but they are *shallow* and does not seem to truly understand
the deep structures of syntax/semantics.
If someone puts a gun to your head and asks you to pick the most promising
approach, what would you say?
To me, it seems that an ideal system is a rule-based one that can learn
statistically.
The innovation in my approach is to open the rules-base to the public.
Some people out there have a much better understanding of the complexities
of NL, and they don't need to be computational linguists to be able to
enter some rules, if the system is easy to use, though some training is
needed.
By the way, for the sub-goal of Basic English, a rule-based system may be
very adequate. And that's useful as a first NL interface.
YKY
-----
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
