Online inference of topics with latent dirichlet allocation

Kevin R. Canini; Lei Shi; Thomas L. Griffiths

Online inference of topics with latent dirichlet allocation

Kevin R. Canini, Lei Shi, Thomas L. Griffiths

Research output: Contribution to journal › Conference article › peer-review

Abstract

Inference algorithms for topic models are typically designed to be run over an entire collection of documents after they have been observed. However, in many applications of these models, the collection grows over time, making it infeasible to run batch algorithms repeatedly. This problem can be addressed by using online algorithms, which update estimates of the topics as each document is observed. We introduce two related Rao-Blackwellized online inference algorithms for the latent Dirichlet allocation (LDA) model incremental Gibbs samplers and particle filters and compare their runtime and performance to that of existing algorithms.

Original language	English (US)
Pages (from-to)	65-72
Number of pages	8
Journal	Journal of Machine Learning Research
Volume	5
State	Published - 2009
Externally published	Yes
Event	12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009 - Clearwater, FL, United States Duration: Apr 16 2009 → Apr 18 2009

All Science Journal Classification (ASJC) codes

Software
Control and Systems Engineering
Statistics and Probability
Artificial Intelligence

Cite this

@article{3e50ef478aba4be3a311ed8094e7f40b,

title = "Online inference of topics with latent dirichlet allocation",

abstract = "Inference algorithms for topic models are typically designed to be run over an entire collection of documents after they have been observed. However, in many applications of these models, the collection grows over time, making it infeasible to run batch algorithms repeatedly. This problem can be addressed by using online algorithms, which update estimates of the topics as each document is observed. We introduce two related Rao-Blackwellized online inference algorithms for the latent Dirichlet allocation (LDA) model incremental Gibbs samplers and particle filters and compare their runtime and performance to that of existing algorithms.",

author = "Canini, {Kevin R.} and Lei Shi and Griffiths, {Thomas L.}",

year = "2009",

language = "English (US)",

volume = "5",

pages = "65--72",

journal = "Journal of Machine Learning Research",

issn = "1532-4435",

publisher = "Microtome Publishing",

note = "12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009 ; Conference date: 16-04-2009 Through 18-04-2009",

}

TY - JOUR

T1 - Online inference of topics with latent dirichlet allocation

AU - Canini, Kevin R.

AU - Shi, Lei

AU - Griffiths, Thomas L.

PY - 2009

Y1 - 2009

N2 - Inference algorithms for topic models are typically designed to be run over an entire collection of documents after they have been observed. However, in many applications of these models, the collection grows over time, making it infeasible to run batch algorithms repeatedly. This problem can be addressed by using online algorithms, which update estimates of the topics as each document is observed. We introduce two related Rao-Blackwellized online inference algorithms for the latent Dirichlet allocation (LDA) model incremental Gibbs samplers and particle filters and compare their runtime and performance to that of existing algorithms.

AB - Inference algorithms for topic models are typically designed to be run over an entire collection of documents after they have been observed. However, in many applications of these models, the collection grows over time, making it infeasible to run batch algorithms repeatedly. This problem can be addressed by using online algorithms, which update estimates of the topics as each document is observed. We introduce two related Rao-Blackwellized online inference algorithms for the latent Dirichlet allocation (LDA) model incremental Gibbs samplers and particle filters and compare their runtime and performance to that of existing algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84862273721&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862273721&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84862273721

SN - 1532-4435

VL - 5

SP - 65

EP - 72

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

T2 - 12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009

Y2 - 16 April 2009 through 18 April 2009

ER -

Online inference of topics with latent dirichlet allocation

Abstract

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this