Data analysis with Yenman diagrams

I use that term, sounds more official.

For the economist, just grab lists and lists of data with which to construct an X axis. Put them all in the sequence they came. Successive bot transformations on the data, 2 at a time. Feed their output to another layer of bot, 2 at a time. Let the build graph build itself, add bots as needed. The encoding graph end up as a polynomial in integer exponent with 1/N coefficient. They coil in the graph, recursive sets.  For example, they should be able to pick out city structure with a fair sample of stats, as a short order polynomial.. The graph should always adapt to the data. The graph will break when the data changes structure, adiabatic change, then reconnect.

Probably already working fine at Goldman Sachs where they do their activity model.  It generates the adapting production function based on minimizing exchanges. But the most obvious result will be city structure, then large state, then the rest. What would be DC structure? Isolated graph, almost certainly, sorry for the Kanosians. But government is long term, up the encoder with a lousy cotangent function. City structure will dominate.  Add in housing data and try to prove something, like some adaptation of the graph.