Multivariate analysis; Stratified sets; Bayesian statistical decision theory
Data mining, at times, involves applying cluster analysis to large databases
containing multivariate normal parameters. Most finite mixture clustering methods use
the expectation-maximum (EM) algorithm to find solutions to the unknown parameters,...
In this thesis, support vector machines, neural networks, logistic regression, naïve Bayes, classification and regression trees, the C5.0 algorithms, QUEST, CHAID and discriminant analysis have been implemented on nine real-world datasets. The...
