SUO: Re: Architecture of an intelligent ontology development algorithm
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Rich,
A few last thoughts. The literature is vast. Just in AI,
I have counted 3 or 4 revivals of the abduction issue,
starting with McCulloch at the very upstart, who had
read Peirce deeply, I forget the names attached to
the next revival, then Josephson once or twice.
The main caution in reading is to be sure it's
really abductive reasoning and not some other
duction. Claims like "abduction is nothing but
Bayesian inference" or "abduction is nothing but
inverse resolution" are nothing but ignoratios.
Standard advice ahead -- Read more Peirce.
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Richard Cooper wrote:
>
> Jon Awbrey wrote:
> > JA = Jon Awbrey
> > RC = Richard Cooper
> >
> > RC: Jon, I still like the advice you gave a few iterations ago,
> > so I snipped it out of some recent emails and reapplied it
> > as below:
> >
> > JA: I would have thought that a fairer summary of what's been
> > said here all summer long on many different threads is that there
> > is really no such thing as a hypothesis-free algorithm for discovery --
> > actually, that's more like a summary of what's been discovered about
> > discovery over the last few thousand summers, but who's counting?
> > So I think that a better question might be something along the
> > following lines:
> >
> > JA: How are concept-driven (analytic, axiomatic, rationalist,
> > top-down) methods
> > and data-driven (synthetic, contingent, empiricist,
> > bottom-up) procedures
> > best to be integrated in human inquiry, or in the
> > reconstitutions thereof,
> > given that the distinction between analytic and synthetic
> > is more relational,
> > interpretive, or "situated" than it is absolute,
> > invariant, or "essential"?
> >
> > RC: This succinctly states the core idea of how an automated or even
> > computer-aided tool could proceed to generate useful ontologies.
> >
> > RC: Hypothesis formation is based on learned conceptualizations
> > from past experience. I'm still reading your refernce to
> > "anticipatory" systems:
> >
> >
> http://www.anticipation.info/texte/rosen/anticipatorysystems_rosen.pdf
>
> Did I give you that? I meant to give you Mihai Nadin's home page:
>
> http://www.code.uni-wuppertal.de/welcome.shtml
>
> RC on JA: You did; I traced the links until I found a paper called "Introduction"
> or "tutorial" or something overviewish like that.
>
> RC: but I like the general tone of a faster-than-real-time model M guiding a
> real time situated system S in modifying its behavior. However, the linear
> system analogy is a bit antiquated, and the term "hypothesis formation" (or
> "abduction" as you Peirceans prefer) isn't conceptually compatible with
> a linear system model.
>
> RC: I like the GA and GP models better. Using symbolic expressions
> rather than linear projections seems more intuitive and natural
> for the areas I like to concentrate on. Forming new expressions
> from combinations of older expressions makes it possible to trace
> the formation of hypotheses to experiences, much like requirements
> can be made traceable to designs using a graph model. So a GP model
> is probably most natural.
>
> Many people like these things, but I tend to shy away from flip-a-coin methods
> for much the same reason that I don't buy theorem provers that say "Yup, it's true
> -- but I can't tell you why". It's the understanding that I'm after, not the simple
> reassurance.
>
> RC on JA: Agreed. That's why I like explainable expressions of some type,
> whether S-expressions or IDEF0 activity models, that show the structure
> and meaning of the conjoined and disjoined concepts. It all has to be
> put into clear terms to be understandable and conveyable.
>
> RC: Then there's the issue of where data comes from to drive the formation of hypotheses.
> For an automated system to reach "good" hypotheses, it should have "good" data --
> data from naturally occuring intelligences. So I subscribe to the idea of
> data mining of English text that has been well written, well annotated,
> and reviewed by various interested parties.
>
> Sigh, if only we could get Nature to write better.
>
> RC on JA: Nature won't cooperate. But you can get great literature from
> lots of sources. Consider Elmore Leonard's excellent prose and full
> action novels. That, annotated by a few "experts" who think they understand
> what each agent is thinking, wanting, accomplishing, and reconceptualizing
> in each sentence, paragraph and chapter. Togther, these two sources
> would provide the manna for producing conceptual honey.
>
> Also, consider the biological research literature. There's too much of
> it for research biologists to read. Its all well written and edited,
> selected for importance, and full of wise judgements. Annotated for
> processing purposes, this is good grist for a bioontological project.
>
> RC: It seems to me that the nominal task of the SUO group could be best
> addressed in producing such a tool (automated, or computer-aided, ...)
> to browse English sentences, data models, UML designs, CGs, or other data
> we would consider useful for the purpose. At least that approach avoids the
> legislative extremes, because we can argue about what data to use, what methods
> of representation, and how the system should function instead of what the ontology
> itself is.
>
> RC: We seem to be running dry on ontologies themselves.
>
> But you're asking for oceans just to prime the pump ...
> and then all you'd get is salt water anyway ...
>
> RC on JA: I don't think so. It takes a scientist to do science. We have
> to filter out the salt water and distill the important parts from the
> waste products.
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o