Artificial intelligence has advanced to the point
where AI Minds can engage in machine reasoning by
inference. For example, if we click on the
Artificial General Intelligence (AGI) link
http://www.scn.org/~mentifex/AiMind.html
with Microsoft Internet Explorer (MSIE) and
we enter a statement like "boys are kids",
we obtain the following conversation from
an actual transcript:
AiMind version 21mar13A on Fri Mar 22 10:01:59 PDT 2013
Robot: I HELP KIDS
Human:
Robot:
Human: boys are kids
Robot: DO BOYS MAKE ROBOTS
Human: no
Robot: KIDS MAKE ROBOTS
Human:
Robot:
Human: eve is a woman
Robot: DO EVE HAVE CHILD
Human: no
Robot: WOMEN HAVE CHILD
Human:
Robot:
Human: eve knows boys
Robot: BOYS DO NOT MAKE ROBOTS
Human:
As you can see in the AI Mind Transcript above,
the artificial intelligence knows a few things
about kids ("Kids make robots") and about women
("Women have a child"). When you tell the AI
that boys are kids or that Eve is a woman,
the AI reasons that perhaps boys make robots
or perhaps Eve has a child. The AI asks you,
the human user, to confirm or refute each inference.
If you input "no" to refute the inference, the AI
Mind retroactively adjusts its knowledge base (KB)
and stores the negation of the inference.
The German artificial intelligence at
http://www.scn.org/~mentifex/DeKi.txt
can reason in German with inference.
Mentifex (Arthur)
--
http://www.tfeb.org/lisp/mad-people.html
http://code.google.com/p/mindforth/wiki/InFerence
http://www.amazon.com/The-Art-Meme-ebook/dp/B007ZI66FS
http://www.chatbots.org/ai_zone/viewthread/1236/#13068