Re: AIML and finite state machines

 

The Emotional I Ching is currently on line
(http://members.iimetro.com.au/~lofting/Emotional/homep.html ). It allows
for accessing emotional assessments of a situation and conversion to some
other form of assessment - all through the use of questions. It is an
example of a "Language of the Vague". How it works is covered in the FAQ on
the site, more formal perspective is covered in the Abstract Domain Model
page http://members.iimetro.com.au/~lofting/myweb/AbstractDomain.html with a
focus on an upper ontology. The EIC book will be out soon.

I am converting the EIC format into an AIML model as a set of tags to be
plugged-in - the issue is in refinements differentiating primary from
secondary emotions. The IDM abstract domain focus in on differentiating the
neural hierarchy where BENEATH the layer of symbol creation and so multiple
context identifications is a level of meaning that is label-free in that it
covers a SINGLE context. Since the labels levels are dependent upon this
base level for their operation so we can translate a set of labels from one
specialist domain to those of another by exploiting the sameness present in
the pre-labels layer.

The IDM abstract domain model identifies the classes of meaning derived in
this single context realm where all labels will point to these classes. Thus
we take specialist labels, convert them to the sameness level classes and
the back up into a different set of specialist labels. In the EIC we use
fight/flight derived classes to translate into yang/yin derived classes.

At this level of no labels, and so single context, it is the HIERARCHY that
elicits meaning where this is in the form of applying a class of dichotomy
recursively and so moving from general to particular and creating classes of
possible meanings. Selection of elements of the dichotomy, through
questions, then maps out a path of actuals through the set of potentials to
give us the specific current meaning.

There are three classes of dichotomy - symmetric (Equivalence focus),
anti-symmetric (traditional Aristotle XOR focus), and asymmetric (Part/whole
processing, also IMP focus). What is of interest is that the initial
recursive process is mechanistic but given some depth (at least 6 levels) we
see emerge the organic in the form of literal categories suddenly being able
to be used figuratively (efficiently so, less levels lack such) and so as
sources of analogy that is 'hard coded' into the methodology of recursion.
At this level we can get each class to have its properties and methods
described by analogy to all of the other classes - as such we have here the
beginnings of languages.

Chris
http://members.iimetro.com.au/~lofting/myweb/introIDM.html


_______________________________________________
This is the alicebot-general mailing list
Reply to alicebot-general@...
Unsubscribe and change preferences at http://list.alicebot.org/mailman/listinfo/alicebot-general
Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
Learn to read at http://www.literacy.org/