Letters: An LVQ-based adaptive algorithm for learning from very small codebooks

Authors:
J. S. Sánchez;A. I. Marqués
Affiliations:
Dept. de Llenguatges i Sistemes Informítics, Universitat Jaume I, Av. Sos Baynat s/n, E-12071 Castelló de la Plana, Spain;Dept. d'Administració d'Empreses i Mírqueting, Universitat Jaume I, Av. Sos Baynat s/n, E-12071 Castelló de la Plana, Spain
Venue:
Neurocomputing
Year:
2006

Citing 5
Cited 3

A new definition of neighborhood of a point in multi-dimensional space

Pattern Recognition Letters
Self-organizing maps

Self-organizing maps
On the use of neighbourhood-based non-parametric classifiers

Pattern Recognition Letters - special issue on pattern recognition in practice V
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
A study of cross-validation and bootstrap for accuracy estimation and model selection

IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2

Rough-fuzzy weighted k-nearest leader classifier for large data sets

Pattern Recognition
A novel template reduction approach for the K-nearest neighbor method

IEEE Transactions on Neural Networks
Optimized Learning Vector Quantization Classifier with an Adaptive Euclidean Distance

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part I

Quantified Score

Hi-index	0.01

Visualization

Abstract

The present paper introduces an adaptive algorithm for competitive training of a nearest neighbor (NN) classifier when using a very small codebook. The new learning rule is based on the well-known LVQ method, and uses an alternative neighborhood concept to estimate optimal locations of the codebook vectors. Experiments over synthetic and real databases suggest the advantages of the learning technique here introduced.