File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Joint topic-document modeling via low-dimensional sparse models

TitleJoint topic-document modeling via low-dimensional sparse models
Authors
Min, KeruiMa, Yi
Keywordsnon-negative matrix factorization
topic modeling
Issue Date2013
Citation
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2013, p. 8590-8594 How to Cite?
DOI: http://dx.doi.org/10.1109/ICASSP.2013.6639342
AbstractTopic modeling is a well-known approach for document analysis. In this paper, we propose a new model, and corresponding optimization algorithm for topic modeling. Experimental results on polarity classification demonstrate that the new model provides a more accurate characterization for document corpus, and archived higher classification accuracy compared to Latent Dirichlet Allocation (LDA). © 2013 IEEE.
Persistent Identifierhttp://hdl.handle.net/10722/326967
ISSN
1520-6149

 

DC FieldValueLanguage
dc.contributor.authorMin, Kerui-
dc.contributor.authorMa, Yi-
dc.date.accessioned2023-03-31T05:27:50Z-
dc.date.available2023-03-31T05:27:50Z-
dc.date.issued2013-
dc.identifier.citationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2013, p. 8590-8594-
dc.identifier.issn1520-6149-
dc.identifier.urihttp://hdl.handle.net/10722/326967-
dc.description.abstractTopic modeling is a well-known approach for document analysis. In this paper, we propose a new model, and corresponding optimization algorithm for topic modeling. Experimental results on polarity classification demonstrate that the new model provides a more accurate characterization for document corpus, and archived higher classification accuracy compared to Latent Dirichlet Allocation (LDA). © 2013 IEEE.-
dc.languageeng-
dc.relation.ispartofICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings-
dc.subjectnon-negative matrix factorization-
dc.subjecttopic modeling-
dc.titleJoint topic-document modeling via low-dimensional sparse models-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/ICASSP.2013.6639342-
dc.identifier.scopuseid_2-s2.0-84890445570-
dc.identifier.spage8590-
dc.identifier.epage8594-

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats