A Bayesian mixture model for clustering and selection of feature occurrence rates under mean constraints

Statistical Analysis and Data Mining

Reich, Brian J.¹  Bondell, Howard D.¹  Li, Qiwei²  Vannucci, Marina²  Guindani, Michele³ 

[1] North Carolina State University Department of Statistics Raleigh North Carolina;Rice University Department of Statistics Houston Texas;University of California Department of Statistics Irvine California

关键词: Bayesian nonparametrics;    count data;    feature selection;    Poisson mixture;    text analysis;   

DOI  :  10.1002/sam.11350

学科分类：社会科学、人文和艺术（综合）

来源: John Wiley & Sons, Inc.

【 摘 要 】

In this paper, we consider the problem of modeling a matrix of count data, where multiple features are observed as counts over a number of samples. Due to the nature of the data generating mechanism, such data are often characterized by a high number of zeros and overdispersion. In order to take into account the skewness and heterogeneity of the data, some type of normalization and regularization is necessary for conducting inference on the occurrences of features across samples. We propose a zero-inflated Poisson mixture modeling framework that incorporates a model-based normalization through prior distributions with mean constraints, as well as a feature selection mechanism, which allows us to identify a parsimonious set of discriminatory features, and simultaneously cluster the samples into homogenous groups. We show how our approach improves on the accuracy of the clustering with respect to more standard approaches for the analysis of count data, by means of a simulation study and an application to a bag-of-words benchmark data set, where the features are represented by the frequencies of occurrence of each word.

【 授权许可】

Unknown