|InFerence Strong AI E-book Wins Favorable Mentifex Review||ment...@myuw.net||10/22/13 7:54 AM|
Logic dictates that free, open-source Strong AI
(artificial intelligence) shall disperse across
the Web and evolve into the brains and minds of
intelligent, autonomous robots. These AI Minds
are able to think with automated reasoning by
using the logical InFerence module, which infers
new knowledge from old knowledge by reasoning
from the human verbal input of new ideas which
assert that an entity is an instance of a predicate
class of all such entities. For instance, a user
might input the new knowledge, "Polly is a bird."
The InFerence module searches the AI memory for
previous records of ideas where the predicate class
was used in the plural number, such as "birds," to
make a general statement about the predicate class,
such as, "Birds have wings." If the InFerence module
finds such prior knowledge, the AI makes a silent,
not yet confirmed inference that the new subject,
such as "Polly," being a bird, may have wings.
The InFerence module then calls the AskUser module
to ask the human user a yes-or-no question,
"Does Polly have wings?" If the human user
answers "yes," then the automated reasoning
confirms the inference as a chain of concepts
linked together by associative tags to form a
thought as an item of knowledge. If the human
user answers "no," then the KbRetro module
retroactively adjusts the knowledge base (KB)
of the AI Mind to negate the inference and to leave
a negated chain of concepts recording the negated idea,
"Polly does not have wings." If the human user
does not answer "yes" or "no" but answers something
like "perhaps" or makes no response at all, the AI
abandons the inference as an unconfirmed idea and
continues thinking about whatever matters come to mind.
http://www.nlg-wiki.org/systems/Mind reasons in English.
http://www.nlg-wiki.org/systems/Wotan reasons in German.
http://www.nlg-wiki.org/systems/Mind.Forth is in English.
http://www.nlg-wiki.org/systems/Dushka thinks in Russian.
"InFerence" is an e-book on automated reasoning at
and published by Amazon across the universe:
http://www.amazon.co.uk/dp/B00FKJY1WY (United Kingdom)
http://www.amazon.com/dp/B00FKJY1WY (United States)