Prototype optimization based on minimizing classification oriented error function

A novel method of constructing optimized prototypes for nearest-neighbor cl assification is proposed. Based on an effective classification oriented err or function containing class identification and class separation components , the corresponding updating rules for prototypes and feature weights are d erived. By minimizing the error function, the optimized prototypes and feat ure weights from the nearest-neighbor classification point of view can be e ffectively constructed. The proposed method consists of several distinct pr ocess; Second, multiple prototypes not belonging to the true class of input sample x are updated when x is classified incorrectly; Third, it intrinsic ally assigns different learning factors to training samples, which enables a large amount of learning from constructive samples, and limited learning from outlier ones; Fourth, by adding a class separation component it avoids the degenerated situation where different prototypes coincide at the same feature position. Experiment results have shown that the proposed methods s uperior to LVQ2 and other method in previous work (Huang, 1995).