Autonomous and Deterministic Probabilistic Neural Network Using Global k-Means
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.
KeywordsProbabilistic Neural Network Cluster Centroid Correct Classification Rate Dimension Size Random Initialization
Unable to display preview. Download preview PDF.
- 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.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.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.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