[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: a propos of artificial intelligence, linguistics, and geckos

On Thu, 20 Jan 2000 scozens@pwj.co.jp wrote:
>      Communication between components: when trying to answer
> a question, we may need to head off to a web search component
> to find information about the topic, which may in turn fire
> off a dictionary component to help with understanding something
> on the returned web page, and so on.

I thought this was sticky, but...

> We're also seeing the communication aspect coming up, which is
> an important issue. (Are we starting to model social patterns?)
> What do *you* do when you need to find something out? You
> either already know it (it's in your DBM, as it were) or you
> can ask someone else, you can look it up on the web or in an
> encyclopaedia. In some situations, you may think you've got the
> answer, but it doesn't seem to fit the question somehow. A
> human can tell pretty easily when he's got the wrong reference
> source for a topic; the query has a series of mental
> associations. When we're looking up quavers, we can usually
> work out if the answer's supposed to relate to music or cheesy
> potato chips. We're about to hit a really nasty knowledge
> representation issue here.


  I'm only going to comment on the first one... I started thinking about
information validity.  It sounds like the path is being paved to a much
smarter AI, and information validity will be a key component.  Everyone
who's run an infobot knows how much invalid data is scraped from IRC, so
I see these main topics:

  How and where to get and accept information.

  How to process and reference, and relate information.

  How to store new information for future refence.

  How to detect and remove bogus information that's already stored.

In addition to those, like Simon said, "The ability to add a filter
to any stage of input/output." is essential.

I have a feeling I've rehashed something from an AI 101 course... But do
we really want to dive this deep into it?  Or keep it simple with the
absolute need for human interaction with the "forget" ability? Or perhaps
give it a hard-coded fact-base, and build "soft" information off of what
is determind as fact?  (i.e. no on-the-fly dictionary lookups)  Perhaps
though if it starts to understand individual words, a thesauraus may be of
some help to point towards the right song about cheesy potato chips.

This is the batter for making a nice cake anyway.  ...I think I'm gonna be
spending my extra time looking into other batters to see how other chefs
cook, since most of my AI and linguistics has been gleemed from cartoons
and the discovery channel.