Tuesday, February 19, 2008

Analysis via context migration, this to define the way composable information units build structured information space

Ok doc mud, let's adopt a problem context migration problem solving approach. Let's adopt a competitive space, for genetics evolution. It can be important to define some minor points.

Information units are bees of a virtual swarm: [info unit] -> [DTO for data linking and association] -> [runnable DTO ?] -> [bee]

All the information units created in a shot makes a chain and this is a generation.

The fact that in a shot all the information units (-> bees) have been created at the same shot time "t" is extremely relevant. This says that the approach of considering timed shots capturing is coherent with the rest of te theory.

Let's go to see your post doc mud, but the fusion of genetic programming on a virtualized swarm computing model needs more care.

1 comment:

Vincenzo said...

Since you got this astonishingly flattering nickname for me I will call you henceforward Dr.B.Dover.
You probably misunderstood my thoughts. The population is not composed by single information units. This would break the correlation between events (both in the same snapshot AND in different instants).
The population is composed by chains of information units.
Let's try to explain this more carefully with an example:
- 3 bumpers (left, forward,right) with status 1/0
- 1 temperature sensor (float value with a fixed range)
- 3 proximity sensors (left,forward,right)

The condition for the quality of the system is that the bumpers are never on (i.e.the car never touches anything)

of all the possible chains the only meaning are the ones involving the bumpers AND the proximity sensors.

The only surviving chains are the ones in which the proximity sensors give a costant (or not decreasing) value (i.e. the car is not approaching anything).

This gives you both the inference rules for the expert system (which we can apply from now on) AND the correlation between data.

What happens with more the one Car is that we can consider the population of ALL (or a random set) the chains ignoring the "affinity" between the info units (i.e. the ones belonging to the same car)
we can introduce some "affinity" rules to reduce the population number.

of all the possible chains