[Fis] Self-structurizing systems

Krassimir Markov itheaiss at gmail.com
Sun Jan 7 00:32:04 CET 2024


Dear colleagues,

First, the feeling of agnosticism in the commented third transition does
not leave me. I hope I'm wrong...

Second, many thanks for the interesting opinions and thoughts.
For me, they are a good reason to go back about 15 years to some projects I
was working on back then. The main driver and generator of ideas was Prof.
Viktor Gladun. I do not have the opportunity here to tell and present all
the models that have been built until working systems and practical results
are reached.
Therefore, I will say a few words about the possibility to discover new
subsets of causal features of objects that constitute affordances that can
be seized by heritable variation and natural selection.
Yes, Category Theory can be used in the initial stages of task analysis to
achieve a collective understanding of the problem and possible approaches
to solving it.
Work then moves into the areas of classification, clustering, pattern
recognition, prediction, feature processing, transformations, data mining,
and knowledge discovery.
It has interesting results and will probably interest you. It's about
self-structurizing systems like we're talking about now. This is what Prof.
Gladun describes in an article published in ITHEA's scientific journal.
There's an example there that's similar to Stu's task, but it's about
predicting new properties of new chemical compounds. Here is a link to the
article: https://urldefense.com/v3/__http://www.foibg.com/ijita/vol15/ijita15-1-p01.pdf__;!!D9dNQwwGXtA!RqsoCnAuTfddfrbGgU9SCAdwn-hINg9s3zygP1snCzMqLErLrrnoGW1Cx6AL22sm3P-Pa47j8BeS4OySyZw$ 
In conclusion, I will note that it is normal to believe in the
impossibility of achieving something until it is achieved by someone who
does not know that it could not be achieved.
With respect,
Krassimir
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20240107/a999aed9/attachment.html>


More information about the Fis mailing list