A new sequential classifier using information criterion window

摘要

We propose a model-based parametric unsupervised sequential classifier with an information criterion window (ICW), based on the assumption that the number of clusters changes smoothly from a stage to the following stage, to determine the number of clusters at each stage and to perform classification on the observed mixture data. The proposed approach provides possible solutions for two problems of pattern recognition: adaptive classification and sequential clustering validation. Simulation results demonstrate that the proposed algorithm provides good results for pure mixed data and successfully performs clustering for the mixed Gaussian data in the large-sample limit.