This paper is concerned about studying modeling-based methods in cluster\nanalysis to classify data elements into clusters and thus dealing with time series\nin view of this classification to choose the appropriate mixed model. The\nmixture-model cluster analysis technique under different covariance structures\nof the component densities is presented. This model is used to capture\nthe compactness, orientation, shape, and the volume of component clusters in\none expert system to handle Gaussian high dimensional heterogeneous data\nset. To achieve flexibility in currently practiced cluster analysis techniques.\nThe Expectation-Maximization (EM) algorithm is considered to estimate the\nparameter of the covariance matrix. To judge the goodness of the models,\nsome criteria are used. These criteria are for the covariance matrix produced\nby the simulation. These models have not been tackled in previous studies.\nThe results showed the superiority criterion ICOMP PEU to other criteria.\nThis is in addition to the success of the model based on Gaussian clusters in\nthe prediction by using covariance matrices used in this study. The study also\nfound the possibility of determining the optimal number of clusters by\nchoosing the number of clusters corresponding to lower values for the different\ncriteria used in the study.
Loading....