Autonomous and Deterministic Probabilistic Neural Network Using Global k-Means

  • Roy Kwang Yang Chang
  • Chu Kiong Loo
  • Machavaram V. C. Rao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3971)


We present a comparative study between Expectation-Maximization (EM) trained probabilistic neural network (PNN) with random initialization and with initialization from Global k-means. To make the results more comprehensive, the algorithm was tested on both homoscedastic and heteroscedastic PNNs. Normally, user have to define the number of clusters through trial and error method, which makes random initialization to be of stochastic nature. Global k-means was chosen as the initialization method because it can autonomously find the number of clusters using a selection criterion and can provide deterministic clustering results. The proposed algorithm was tested on benchmark datasets and real world data from the cooling water system in a power plant.


Probabilistic Neural Network Cluster Centroid Correct Classification Rate Dimension Size Random Initialization 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Specht, D.F.: Probabilistic Neural Network. Neural Networks 3, 109–118 (1990)CrossRefGoogle Scholar
  2. 2.
    Likas, A., Vlassis, N., Verbeek, J.J.: The Global k-means Clustering Algorithm. Pattern Recognition 36, 451–461 (2003)CrossRefGoogle Scholar
  3. 3.
    Wu, C.: On the Convergence Properties of the EM Algorithm. Annals Statistics 11, 95–103 (1983)MATHCrossRefGoogle Scholar
  4. 4.
    Xu, L., Jordan, M.I.: On Convergence Properties of the EM Algorithm for Gaussian Mixtures. Neural Computation 8, 129–151 (1996)CrossRefGoogle Scholar
  5. 5.
    Yang, Z.R., Chen, S.: Robust Maximum Likelihood Training of Heteroscedastic Probabilistic Neural Networks. Neural Networks 11, 739–747 (1998)CrossRefGoogle Scholar
  6. 6.
    Ordonez, C., Omiecinski, E.: FREM: Fast and Robust EM Clustering for Large Data Sets. In: Proc. of the 11th CIKM, Clustering Algorithms., pp. 590–599. ACM Press, New York (2002)Google Scholar
  7. 7.
    Blake, C.L., Merz, C.J.: UCI Repository of Machine Learning Databases. Department of Information and Computer Sciences. University of California, Irvine (1998)Google Scholar
  8. 8.
    Zarndt, F.: A Comprehensive Case Study: An Examination of Machine Learning and Connectionist Algorithms. MSc Thesis. Dept. of Computer Science, Brigham Young University (1995)Google Scholar
  9. 9.
    Ster, B., Dobnikar, A.: Neural Networks in Medical Diagnosis: Comparison with Other Methods. In: Bulsari, A., et al. (eds.) Proc. Int. Conference EANN 1996. Solving Engineering Problems with Neural Networks, pp. 427–430. Systems Engineering Association, Turku (1996)Google Scholar
  10. 10.
    Chen, K.Y., Lim, C.P., Lai, W.K.: Application of a Neural Fuzzy System with Rule Extraction to Fault Detection and Diagnosis. Journal of Intelligent Manufacturing 16, 679–691 (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Roy Kwang Yang Chang
    • 1
  • Chu Kiong Loo
    • 2
  • Machavaram V. C. Rao
    • 2
  1. 1.Faculty of Information Science and TechnologyMutilmedia UniversityMelakaMalaysia
  2. 2.Faculty of Engineering TechnologyMutilmedia UniversityMelakaMalaysia

Personalised recommendations