Improving topic coherence with latent feature word representations in MAP estimation for topic modeling

Dat Quoc Nguyen, Kairit Sirts, Mark Johnson

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

Probabilistic topic models are widely used to discover latent topics in document collections, while latent feature word vectors have been used to obtain high performance in many natural language processing (NLP) tasks. In this paper, we present a new approach by incorporating word vectors to directly optimize the maximum a posteriori (MAP) estimation in a topic model. Preliminary results show that the word vectors induced from the experimental corpus can be used to improve the assignments of topics to words.
Original languageEnglish
Title of host publicationProceedings of the Australasian Language Technology Association Workshop 2015
EditorsBen Hachey, Kellie Webster
Place of PublicationStroudsburg, PA
PublisherAssociation for Computational Linguistics (ACL)
Pages116-121
Number of pages6
Volume13
Publication statusPublished - 1 Dec 2015
EventAustralasian Language Technology Association Workshop (13th : 2015) - Parramatta, Australia
Duration: 8 Dec 20159 Dec 2015

Conference

ConferenceAustralasian Language Technology Association Workshop (13th : 2015)
Abbreviated titleALTA 2015
CountryAustralia
CityParramatta
Period8/12/159/12/15

Keywords

  • MAP estimation
  • LDA
  • topic model
  • word vectors
  • topic coherence

Fingerprint Dive into the research topics of 'Improving topic coherence with latent feature word representations in MAP estimation for topic modeling'. Together they form a unique fingerprint.

Cite this