Re: topological and algebraic structures over strings and graphs



sid myers wrote:

On Jul 26, 3:27 pm, galathaea <galath...@xxxxxxxxxx>
wrote:
there have been a flurry of posts recently on graph
problems
and their relation to the question of whether
p=np

also
i have recently been looking back upon a
simplistic model of relative geometry
based loosely on distance geometry with some
derived "flat" dynamics
to see if i could develop some generalisations or
furthur results
and the formulation has been in terms of certain
structures over graphs
that may be useful in some of the other graph
pursuits

so i thought i'd post some ramblings and see if i
can elicit some references

-+-+-

take a graph G with the chain topology (chains as
open sets)
ie. vertex x is in a neighborhood of vertex y
iff thereExists a walk P with x, y e
verticesOf(P)

this topology preserves the notion of graph
components

last year a paper by alain bretto, alain faisant,
and thierry vallée
"compatible topologies on graphs " in the great
journal "theoretical computer science"
showed the graph isomorphism problem is polynomial
time equivalent to the topological homeomorphism
problem
by analysing a class of compatible topologies
( topologies that preserve connectedness of
vertex subsets )

the chain topology is a compatible topology

-+++-

kontsevich has given an amazing set of homological
correspondences

he has shown that

H (lie algebra) = H (graph complex) = H (group)
* * *

http://arxiv.org/pdf/math/0211464

there appears to be much computational information
here
but i'm not sure how one pulls it off...

i need to
- define explicitly how p=np is stated in the
language of algorithms over graphs
and detail the interpretations of those
structures in the topology of it's dynamics
- provide a means of specifying graph problems
inside the structure of the first step
- define the asymptotic measure to define time
complexity of queries

probably most important to focus on the subgraph
isomorphism here?
or are there other natural problems to interpret?
( i consider the clique and independent set
problems dependent here )

now after reading some interesting associations
that allow interpretation in equations of
dynamics

http://arxiv.org/pdf/0704.3525

which fits into the models i have been trying to
learn more about
that model finite collections of existents
each with distance states to other existents
over graph models of connection
generalising distance geometry to noneuclidean
even noncommutative geometries
i became interested in interpreting this in domain
theory and its fixed point theorems
or even
to keep complexity down because i was already
over my head
just a y-combinator

unrolling recursions

lambda expressions over graph data structures

@##$$..&%^^

but it seems it may be easier to get answers on
strings rather than graphs

they are both computationally (turing) complete
and there are natural representations of graphs
in strings
( adjacency and incidence lists )

and strings have a simple topology

how do i interpret the subgraph isomorphism in
strings?

it seems there is a relabled incidence sublist
problem here...

$%#&&^%&$&^&$&%^^&$%^&$&^$%^

anyway
as can be seen and as always
there are a lot of questions here and few answers

where to start?

good references?

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
galathaea: prankster, fablist, magician, liar

i've decided not to make you extinct. how would you
like to be an
honorary man?

This is your first post (Isn't it?) so, congratulations,
we really need scientific and well educated persons as
you are.

Fernando.
.



Relevant Pages

  • Re: similarity scorem between graphs/chemical structures
    ... >> I've a question concerning the similarity between let's say ... >> In particular I'm working with directed graphs that I want to ... When the fingerprint is expressed as a bit string, ... topology of the molecule. ...
    (sci.chem.analytical)
  • Graph Theory and General Topology
    ... After reading a recent letter that has to do with graphs, ... of terms that are used in general (ie: point-set) topology. ... general topology theory for all I know. ...
    (comp.theory)
  • graph
    ... that work with mobile robots and one of the main strategies of ... control is topology of graphs. ...
    (comp.soft-sys.matlab)