This repo contains code and other information for our paper, to be presented at InfoVis '16.
Surprise Maps use Bayesian Surprise in order to weight heatmaps and cholorpleth maps in order to highlight unexpected events.
The paper itself, and all relevant figures, are available in the Paper directory.
Sample code for computing surprise maps is available in the Code directory.
An interactive demo of a sample surprise map is available in the Demo directory, or running in github pages.