Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Validity Study of Kohonen Self-Organizing Maps

  • Published : 2003.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

Self-organizing map (SOM) has been developed mainly by T. Kohonen and his colleagues as a unsupervised learning neural network. Because of its topological ordering property, SOM is known to be very useful in pattern recognition and text information retrieval areas. Recently, data miners use Kohonen´s mapping method frequently in exploratory analyses of large data sets. One problem facing SOM builder is that there exists no sensible criterion for evaluating goodness-of-fit of the map at hand. In this short communication, we propose valid evaluation procedures for the Kohonen SOM of any size. The methods can be used in selecting the best map among several candidates.

Keywords

References

  1. Neural Networks v.15 Statistical tools to assess the reliability of self-organizing maps de Bodt,E.;Cottrell,M.;Verleysen,M. https://doi.org/10.1016/S0893-6080(02)00071-0
  2. Classification(Second Edition) Gordon,A.D.
  3. The Elements of Statistical Learning Hastie,T.;Tibshirani,R.;Friedman,J.
  4. Korean Journal of Applied Statistics v.16 Principal Components Self-Organizing Map PC-SOM Huh,M.H. https://doi.org/10.5351/KJAS.2003.16.2.321
  5. Korean Journal of Applied Statistics v.15 Bayesian learning for Self Organizing Maps Jun,S.H.;Jorn,H.;Hwang,J. https://doi.org/10.5351/KJAS.2002.15.2.251
  6. Neurocomputing v.21 The self-organizing map Kohonen,T. https://doi.org/10.1016/S0925-2312(98)00030-7
  7. Self-Organizing Maps Kohonen,T.
  8. Kohonen Maps Tree structured self-organizing maps Koikkalainen,P.;E.Oja(ed.);Kaski,S.(ed.)
  9. Pattern Recognition and Neural Network Ripley,R.D.