Complexity optimized data clustering by competitive neural networks

Authors:
Joachim Buhmann;Hans Kühnel
Affiliations:
-;-
Venue:
Neural Computation
Year:
1993

Citing 0
Cited 11

An on-line agglomerative clustering method for nonstationary data

Neural Computation
Probabilistic Topic Maps: Navigating through Large Text Collections

IDA '99 Proceedings of the Third International Symposium on Advances in Intelligent Data Analysis
Online Clustering Algorithms for Radar Emitter Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
ProbMap -- A probabilistic approach for mapping large document collections

Intelligent Data Analysis
Learning with preknowledge: Clustering with point and graph matching distance measures

Neural Computation
Hierarchical, unsupervised learning with growing via phase transitions

Neural Computation
A NON-PARAMETRIC APPROACH TO SIMPLICITY CLUSTERING

Applied Artificial Intelligence
A comparative analysis of clustering methodology and application for market segmentation: K-means, SOM and a two-level SOM

ISMIS'06 Proceedings of the 16th international conference on Foundations of Intelligent Systems
Application of a two-level self organizing map for korean online game market segmentation

SSPR'06/SPR'06 Proceedings of the 2006 joint IAPR international conference on Structural, Syntactic, and Statistical Pattern Recognition
A comparison of target customers in asian online game markets: marketing applications of a two-level SOM

ICANN'06 Proceedings of the 16th international conference on Artificial Neural Networks - Volume Part II
Regularized discriminant entropy analysis

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Data clustering is a complex optimization problem with applicationsranging from vision and speech processing to data transmission anddata storage in technical as well as in biological systems. Wediscuss a clustering strategy that explicitly reflects the tradeoffbetween simplicity and precision of a data representation. Theresulting clustering algorithm jointly optimizes distortion errorsand complexity costs. A maximum entropy estimation of theclustering cost function yields an optimal number of clusters,their positions, and their cluster probabilities. Our approachestablishes a unifying framework for different clustering methodslike K-means clustering, fuzzy clustering, entropy constrainedvector quantization, or topological feature maps and competitiveneural networks.