Abstract
The models of recognition systems are described in general and generalized classifiers in particular. The arbitrary metric is considered as a filter in a feature space. It is shown that optimization of the model of recognition system is reduced to optimization of its classifier model. The results of comparing two decision rules based on the distance function from one and k nearest neighbors are presented.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have