caa a22 4500 445871210 CHVBK 20180317145512.0 cr unu---uuuuu 170323e20110701xx s 000 0 eng 10.1007/s10618-010-0198-2 doi (NATIONALLICENCE)springer-10.1007/s10618-010-0198-2 Mixed-membership naive Bayes models [Elektronische Daten] [Hanhuai Shan, Arindam Banerjee] In recent years, mixture models have found widespread usage in discovering latent cluster structure from data. A popular special case of finite mixture models is the family of naive Bayes (NB) models, where the probability of a feature vector factorizes over the features for any given component of the mixture. Despite their popularity, naive Bayes models do not allow data points to belong to different component clusters with varying degrees, i.e., mixed memberships, which puts a restriction on their modeling ability. In this paper, we propose mixed-membership naive Bayes (MMNB) models. On one hand, MMNB can be viewed as a generalization of NB by putting a Dirichlet prior on top to allow mixed memberships. On the other hand, MMNB can also be viewed as a generalization of latent Dirichletallocation (LDA) with the ability to handle heterogeneous feature vectors with different types of features, e.g., real, categorical, etc.. We propose two variational inference algorithms to learn MMNB models. The first one is based on ideas originally used in LDA, and the second one uses substantially fewer variational parameters, leading to a significantly faster algorithm. Further, we extend MMNB/LDA to discriminative mixed-membership models for classification by suitably combining MMNB/LDA with multi-class logistic regression. The efficacy of the proposed mixed-membership models is demonstrated by extensive experiments on several datasets, including UCI benchmarks, recommendation systems, and text datasets. The Author(s), 2010 Naive Bayes nationallicence Latent Dirichletallocation nationallicence Mixed-membership nationallicence Generative models nationallicence Variational inference nationallicence Logistic regression nationallicence Shan Hanhuai Department of Computer Science & Engineering, University of Minnesota, Twin Cities, Minneapolis, MN, USA aut Banerjee Arindam Department of Computer Science & Engineering, University of Minnesota, Twin Cities, Minneapolis, MN, USA aut Data Mining and Knowledge Discovery Springer US; http://www.springer-ny.com 23/1(2011-07-01), 1-62 1384-5810 23:1<1 2011 23 10618 https://doi.org/10.1007/s10618-010-0198-2 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/s10618-010-0198-2 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Shan Hanhuai Department of Computer Science & Engineering, University of Minnesota, Twin Cities, Minneapolis, MN, USA aut NATIONALLICENCE 700 1- Banerjee Arindam Department of Computer Science & Engineering, University of Minnesota, Twin Cities, Minneapolis, MN, USA aut NATIONALLICENCE 773 0- Data Mining and Knowledge Discovery Springer US; http://www.springer-ny.com 23/1(2011-07-01), 1-62 1384-5810 23:1<1 2011 23 10618 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer