Hierarchical Bayesian models for applications in information retrieval

被引:0
作者
Blei, DM [1 ]
Jordan, M [1 ]
Ng, AY [1 ]
机构
[1] Univ Calif Berkeley, Berkeley, CA 94720 USA
来源
BAYESIAN STATISTICS 7 | 2003年
关键词
variational inference methods; hierarchical Bayesian models; empirical Bayes; latent variable models; information retrieval;
D O I
暂无
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
We present a simple hierarchical Bayesian approach to the modeling collections of texts and other large-scale data collections. For text collections, we posit that a document is generated by choosing a random set of multinomial probabilities for a set of possible "topics," and then repeatedly generating words by sampling from the topic mixture. This model is intractable for exact Probabilistic inference, but approximate posterior probabilities and marginal likelihoods can be obtained via fast variational methods. We also present extensions to coupled models for joint text/image data and multiresolution models for topic hierarchies.
引用
收藏
页码:25 / 43
页数:19
相关论文
共 25 条
[1]  
[Anonymous], 2000, C&H TEXT STAT SCI
[2]  
Attias H, 2000, ADV NEUR IN, V12, P209
[3]  
BAEZAYATES RA, 1999, MODERN INFORMATION R
[4]  
BELI D, 2002, UNPUB J MACHINE LEAR
[5]  
BLEI D, 2002, MODELING ANNOTATED D
[6]  
Chen CA, 1996, ONLINE J KNOWL SYN N, V3, P1
[7]  
de Freitas N., 2001, P 17 C UNC ART INT U, P120
[8]   BAYESIAN METHODS FOR CENSORED CATEGORICAL-DATA [J].
DICKEY, JM ;
JIANG, JM ;
KADANE, JB .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1987, 82 (399) :773-781
[10]  
DIEBOLT J, 1994, J ROY STAT SOC B MET, V56, P363