Probabilistic Topic Models

被引：197

作者：

Blei, David ^{[1
]}

Carin, Lawrence ^{[2
]}

Dunson, David ^{[3
]}

机构：

[1] Princeton Univ, Princeton, NJ 08544 USA

[2] Signal Innovat Grp Inc, Durham, NC USA

[3] Duke Univ, Durham, NC 27706 USA

来源：

IEEE SIGNAL PROCESSING MAGAZINE | 2010年 / 27卷 / 06期

关键词：

Analytical models; Data models; Graphical models; Computational modeling; Bayesian methods; Markov processes; INFERENCE;

D O I：

10.1109/MSP.2010.938079

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

In this article, we review probabilistic topic models: graphical models that can be used to summarize a large collection of documents with a smaller number of distributions over words. Those distributions are called Â¿topicsÂ¿ because, when fit to data, they capture the salient themes that run through the collection. We describe both finite-dimensional parametric topic models and their Bayesian nonparametric counterparts, which are based on the hierarchical Dirichlet process (HDP). We discuss two extensions of topic models to time-series dataÂ¿one that lets the topics slowly change over time and one that lets the assumed prevalence of the topics change. Finally, we illustrate the application of topic models to nontext data, summarizing some recent research results in image analysis. © 2010 IEEE.

引用

页码：55 / 65

页数：11