International Association of Computer Science and Information Technology(IACSIT)
Data classification is the basic approach of data mining and Knowledge Discovery in Databases (KDD). In recent years, cloud classifier based on the cloud theory has been proposed. The most difference between cloud classifier and the traditional classifiers was that classified boundary of cloud classifier is fuzzy. Since current research only focus on the one-dimensional cloud generator algorithm, so this paper presents the classification algorithms based on the multidimensional cloud generator. Moreover, to resolve the complexity of classification which was brought by multi-dimension independent samples, the author proposes a method to so1ve the dimensionality reduction problem of multidimensional samples by one-dimensional cloud charts. Finally, the accuracy of cloud classifier is verified by a classification experiment on a texture database.