Hi all, FYI, I've had a Haskell implementation of a support vector machine learning algorithm around for a while and thought some people might find it interesting to look at/use. There are several advantages it has over SVM-Light; these are listed on the web page, but include sample selection and better control over memory usage. More information and documentation is available at: http://www.isi.edu/~hdaume/SVMseq/ Best, - Hal -- Hal Daume III | hdaume@isi.edu "Arrest this man, he talks in maths." | www.isi.edu/~hdaume