{"title":"Self Organizing Mixture Network in Mixture Discriminant Analysis: An Experimental Study","authors":"Nazif \u00c7al\u0131\u015f, Murat Eri\u015fo\u011flu, Hamza Erol, Tayfun Servi","country":null,"institution":"","volume":55,"journal":"International Journal of Mathematical and Computational Sciences","pagesStart":1015,"pagesEnd":1019,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/5583","abstract":"In the recent works related with mixture discriminant\r\nanalysis (MDA), expectation and maximization (EM) algorithm is\r\nused to estimate parameters of Gaussian mixtures. But, initial values\r\nof EM algorithm affect the final parameters- estimates. Also, when\r\nEM algorithm is applied two times, for the same data set, it can be\r\ngive different results for the estimate of parameters and this affect the\r\nclassification accuracy of MDA. Forthcoming this problem, we use\r\nSelf Organizing Mixture Network (SOMN) algorithm to estimate\r\nparameters of Gaussians mixtures in MDA that SOMN is more robust\r\nwhen random the initial values of the parameters are used [5]. We\r\nshow effectiveness of this method on popular simulated waveform\r\ndatasets and real glass data set.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 55, 2011"}