Latent Dirichlet Allocation with topic-in-set knowledge
D. Andrzejewski, and X. Zhu. Proceedings of the NAACL HLT 2009 Workshop on Semi-Supervised Learning for Natural Language Processing, page 43--48. Stroudsburg, PA, USA, Association for Computational Linguistics, (2009)
Abstract
Latent Dirichlet Allocation is an unsupervised graphical model which can discover latent topics in unlabeled data. We propose a mechanism for adding partial supervision, called topic-in-set knowledge, to latent topic modeling. This type of supervision can be used to encourage the recovery of topics which are more relevant to user modeling goals than the topics which would be recovered otherwise. Preliminary experiments on text datasets are presented to demonstrate the potential effectiveness of this method.
Description
Latent Dirichlet Allocation with topic-in-set knowledge
%0 Conference Paper
%1 Andrzejewski:2009:LDA:1621829.1621835
%A Andrzejewski, David
%A Zhu, Xiaojin
%B Proceedings of the NAACL HLT 2009 Workshop on Semi-Supervised Learning for Natural Language Processing
%C Stroudsburg, PA, USA
%D 2009
%I Association for Computational Linguistics
%K background knowledge lda pub
%P 43--48
%T Latent Dirichlet Allocation with topic-in-set knowledge
%U http://dl.acm.org/citation.cfm?id=1621829.1621835
%X Latent Dirichlet Allocation is an unsupervised graphical model which can discover latent topics in unlabeled data. We propose a mechanism for adding partial supervision, called topic-in-set knowledge, to latent topic modeling. This type of supervision can be used to encourage the recovery of topics which are more relevant to user modeling goals than the topics which would be recovered otherwise. Preliminary experiments on text datasets are presented to demonstrate the potential effectiveness of this method.
%@ 978-1-932432-38-1
@inproceedings{Andrzejewski:2009:LDA:1621829.1621835,
abstract = {Latent Dirichlet Allocation is an unsupervised graphical model which can discover latent topics in unlabeled data. We propose a mechanism for adding partial supervision, called topic-in-set knowledge, to latent topic modeling. This type of supervision can be used to encourage the recovery of topics which are more relevant to user modeling goals than the topics which would be recovered otherwise. Preliminary experiments on text datasets are presented to demonstrate the potential effectiveness of this method.},
acmid = {1621835},
added-at = {2013-05-15T15:41:58.000+0200},
address = {Stroudsburg, PA, USA},
author = {Andrzejewski, David and Zhu, Xiaojin},
biburl = {https://www.bibsonomy.org/bibtex/2fcf352d9f5491cc09a516e87a8376480/schwemmlein},
booktitle = {Proceedings of the NAACL HLT 2009 Workshop on Semi-Supervised Learning for Natural Language Processing},
description = {Latent Dirichlet Allocation with topic-in-set knowledge},
interhash = {858c370d61fd4538da0878b79f97fcb1},
intrahash = {fcf352d9f5491cc09a516e87a8376480},
isbn = {978-1-932432-38-1},
keywords = {background knowledge lda pub},
location = {Boulder, Colorado},
numpages = {6},
pages = {43--48},
publisher = {Association for Computational Linguistics},
series = {SemiSupLearn '09},
timestamp = {2013-05-15T15:41:58.000+0200},
title = {Latent Dirichlet Allocation with topic-in-set knowledge},
url = {http://dl.acm.org/citation.cfm?id=1621829.1621835},
year = 2009
}