##Name: Joel D. Rodriguez-Ortiz
##Node-Link Diagram Insights
Insights gained from this graph include:
In general node-link graphs are good for visually observing topologies, and for intuitively understanding algorithms that are normally hard to express mathematically (shuch as shortest paths).
##Adjacency-Matrix View Insights
Note
All-pairs shortest paths have been calculated in the second adjacency matrix
view presented. Distances have been encoded via colors. Bluer distances
represent the magnitude of shorter distances from A to B, and greener ones
represent longer distances. The exception is red which represents the distance
of 1, to verify correctness with the first adjacency matrix and to make it easy
to find adjacent nodes.
In the first matrix, red is a distance of 1 and blue a distance of 0.
Insights gained from this graph include:
Nodes with a larger number of connections have a larger number of shorter routes to other nodes. If you sort the adjacency matrix by number of direct connections you will see the blue squares gather towards the bottom right of the matrix. Nodes with less direct connections filter towards the top left. Nodes with more direct connections filter towards the bottom right.
ME is adjacent to a single state, NH.
Colorado (CO) is adjacent to six other states.
You can't see it now (because it's corrected), but this view showed me an off-by-one error I had from coding up the all-pairs shortest path algorithm from CLRS. I probably woudln't have seen this from the node-link view.
Adjacency matrix views are suited for high information density and are typically well suited for algorithmic calculations.
##Extras