Keywords

radial basis function networks, distance function

Abstract

Radial Basis Function (RBF) networks typically use a distance function designed for numeric attributes, such as Euclidean or city-block distance. This paper presents a heterogeneous distance function which is appropriate for applications with symbolic attributes, numeric attributes, or both. Empirical results on 30 data sets indicate that the heterogeneous distance metric yields significantly improved generalization accuracy over Euclidean distance in most cases involving symbolic attributes.

Original Publication Citation

Wilson, D. R. and Martinez, T. R., "Heterogeneous Radial Basis Function Networks", Proceedings of the ICNN'96 IEEE International Conference on Neural Networks, pp. 263-1267, 1996.

Document Type

Peer-Reviewed Article

Publication Date

1996-06-01

Permanent URL

http://hdl.lib.byu.edu/1877/2424

Publisher

IEEE

Language

English

College

Physical and Mathematical Sciences

Department

Computer Science

Share

COinS