Hello, Wondering if anyone has come across any decent, light-weight, learning, text categorization tools? A webservice would also be fine. The best I could find so far is something called AI:Categorizer which is alpha and open source. I need something reasonable, preferably off site to save resources. The task is automatically categorizing websites based on the text content of the first few pages and the domain. Categorization needs to roughly follow the dmoz hierarchy. If anyone has even a slight clue I'd very much appreciate your input.