Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/37069
標題: On fast supervised learning for normal mixture models with missing information
作者: Lin, T.I.
林宗儀
Lee, J.C.
Ho, H.J.
關鍵字: Bayesian classifier
data augmentation
EM algorithrn
incomplete
features
Rao-Blackwellization
multivariate-t-distribution
bayesian-analysis
posterior distributions
incomplete data
unknown number
inference
components
algorithm
em
期刊/報告no:: Pattern Recognition, Volume 39, Issue 6, Page(s) 1177-1187.
摘要: It is an important research issue to deal with mixture models when missing values occur in the data. In this paper, computational strategies using auxiliary indicator matrices are introduced for efficiently handling mixtures of multivariate normal distributions when the data are missing at random and have an arbitrary missing data pattern, meaning that missing data can occur anywhere. We develop a novel EM algorithm that can dramatically save computation time and be exploited in many applications, such as density estimation, supervised clustering and prediction of missing values. In the aspect of multiple imputations for missing data, we also offer a data augmentation scheme using the Gibbs sampler. Our proposed methodologies are illustrated through some real data sets with varying proportions of missing values. (c) 2006 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
URI: http://hdl.handle.net/11455/37069
ISSN: 0031-3203
文章連結: http://dx.doi.org/10.1016/j.patcog.2005.12.014
Appears in Collections:統計學研究所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.