Abstract
The development of a Bayesian iterative method of clustering on the basis of feature selection allows you to filter the aggregate of cluster-forming characteristics according to their information content and, consequently, produce a "thin" adjustments to the initial partitioning of objects into clusters. The efficiency of the proposed method of feature selection for clustering, expressed in a significant reduction of their number, for modeling decision-making in credit banking technologies.
Keywords
neural network model, clustering, Bayesian approach, selection of characteristics, clusters, and loan technology
Comments
No posts found