Clustering Based on LVQ and a Split and Merge Procedure

Authors:
Fujiki Morii
Affiliations:
Dept. of Information and Computer Sciences, Nara Women's University, Nara, Japan 630-8506
Venue:
Neural Information Processing
Year:
2008

Citing 5
Cited 0

Algorithms for clustering data

Algorithms for clustering data
Self-organizing maps

Self-organizing maps
Pattern Recognition with Fuzzy Objective Function Algorithms

Pattern Recognition with Fuzzy Objective Function Algorithms
Bayesian model search for mixture models based on optimizing variational bounds

Neural Networks
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Although learning vector quantization (LVQ) based on learning concept is a typical clustering method, we cannot necessarily obtain satisfactory classification results for linearly separable data. In this paper, a new clustering method based on LVQ and a split and merge procedure is proposed to realize reliable classification. Introducing a criterion of whether or not there is only one cluster in each class after clustering by LVQ, split subclasses in a class are merged into appropriate neighboring classes except one subclass. And the validity of the classification result is checked. Under several classification experiments, the performance of the proposed method is provided.