A new approach to classification

Abstract

Clustering is a common and important issue, and finite mixture models based on the normal distribution are frequently used to address the problem. In this article, we consider a classification model and build a mixture model around it. A good assessment of the allocation of observations and number of clusters is easily obtained from this approach.

Keywords:

• clustering
• classification
• latent variable
• normal mixture-model
• random histogram

Acknowledgements

Stephen Walker is grateful for the support of PAPIIT grant IN109906, UNAM, México. Ruth Fuentes-García is grateful to the Mexican Mathematical Society -Sofia Kovalevskaia Fund for their financial support in the final stage of this project. Authors are grateful to the editor and the reviewers for their suggestions.