Right now, I'm making a program that calculates two focal models for an earthquake, based on its aftershocks, and tries to compare the models to find out which one is more likely. The thing is, I have no idea how to compare these models (this is the research part!).
It would be really nice if I could, somehow, use evolutionary algorithms to generate ways to compare these models. Fitness would be the difference in the model comparison between the correct and incorrect model, but I have no idea on what to use as the representation.
Putting this on the "ideas to think about" folder...