Clustering of multivariate binary data with dimension reduction via L1-regularized likelihood maximization

PATTERN RECOGNITION

卷:48

Article

Yamamoto, Michio¹  Hayashi, Kenichi² 

[1] Kyoto Univ, Grad Sch Med, Dept Biomed Stat & Bioinformat, Sakyo Ku, Kyoto 6068507, Japan

[2] Osaka Univ, Grad Sch Med, Suita, Osaka 5650871, Japan

关键词: Binary data;    Clustering;    Dimension reduction;    EM algorithm;    Latent class analysis;    Sparsity;   

DOI  :  10.1016/j.patcog.2015.05.026

来源: Elsevier

【 摘 要 】

Clustering methods with dimension reduction have been receiving considerable wide interest in statistics lately and a lot of methods to simultaneously perform clustering and dimension reduction have been proposed. This work presents a novel procedure for simultaneously determining the optimal cluster structure for multivariate binary data and the subspace to represent that cluster structure. The method is based on a finite mixture model of multivariate Bernoulli distributions, and each component is assumed to have a low-dimensional representation of the cluster structure. This method can be considered as an extension of the traditional latent class analysis. Sparsity is introduced to the loading values, which produces the low-dimensional subspace, for enhanced interpretability and more stable extraction of the subspace. An EM-based algorithm is developed to efficiently solve the proposed optimization problem. We demonstrate the effectiveness of the proposed method by applying it to a simulation study and real datasets. (C) 2015 Elsevier Ltd. All rights reserved.

【 授权许可】

Free