> "Vectors of high dimensionality". Is this interesting to J ? See below.
"Feature vectors of high dimensionality" means, in English, "lots of columns." Aka, each column is a "feature" or "dependent variable" fed to the machine learning algorithm. I think in speech problems it's usually around 100. Speech recognition is good stuff, and the practitioners are often sought after hot-dogs who disappear into various spooky hedge funds and government agencies. J would probably be pretty good for doing this, but a lot of libraries and basic doodads would be necessary to get things done. -SL ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
