I feel as if I had just returned from AI-2007 (at least the expenses were paid just recently by the University) and there’s the deadline approaching for AI-2008, again held at Peterhouse College, UK.

In the meantime since the latest meeting with the agriculture people I have collected some ideas and read some literature. The baseline will probably be that I’ll use some of our (un-)supervised learning methods to find a good model of different agriculture data. There are a few data sets that I can probably work with and I will prepare some Matlab code that compares Neural Networks (MLP, RBF) with k-nearest-neighbor clustering, a simple decision tree and (possibly) support vector machines for regression. I am still in the early stages of this work and I will have to figure out some Matlab toolboxes for the learning algorithms I’d like to use. The paper then will be mainly a description of how to find a suitable model and it will show just a small part of the overall process, but an important one.

For reference, there are some links to Matlab toolboxes, without having evaluated them:

Routines for multi-layer perceptrons and radial basis function networks are built-in, as well as the k-nearest-neighbor clustering seems to be. I’ll publish code right here once the experiments are done. There’s also a good collection of links regarding support vector machines here.