Hi,
I was in the process of writing a package for managing and classifying
feature vectors in LISP, when I decided it might be a good idea if I first
checked if there was such a beast. Ideally, I would like to declare some
class of objects as features, indicate a learning algorithm(s) to use over
the vectors that represent my data, call a function whenever I came across
a feature to indicate its existence, and when I am ready run the particular
supervised/unsupervised learning algorithms I would like to use.
I know it sounds like a stretch, but I am fairly confident it can be done,
so I wanted to know if anyone else has done it before me.
Sunil