Topic models with topic ordering regularities for topic segmentation

Lan Du, John K. Pate, Mark Johnson

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

2 Citations (Scopus)


Documents from the same domain usually discuss similar topics in a similar order. In this paper we present new ordering-based topic models that use generalised Mallows models to capture this regularity to constrain topic assignments. Specifically, these new models assume that there is a canonical topic ordering shared amongst documents from the same domain, and each document-specific topic ordering is allowed to vary from the canonical topic ordering. Instead of full orderings over a set of all possible topics covered by a domain, we make use of top-t orderings via a multistage ranking process. We show how to reformulate the new models so that a point-wise sampling algorithm from the Bayesian word segmentation literature can be used for posterior inference. Experimental results on several document collections with different properties show that our model performs much better than the other topic ordering-based models, and competitively with other state-of-the-art topic segmentation models.

Original languageEnglish
Title of host publicationProceedings - 14th IEEE International Conference on Data Mining, ICDM 2014
EditorsRavi Kumar, Hannu Toivonen, Jian Pei, Joshua Zhexue Huang, Xindong Wu
Place of PublicationPiscataway, NJ
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages6
Publication statusPublished - 2014
EventIEEE International Conference on Data Mining (14th : 2014) - Shenzhen, China
Duration: 14 Dec 201417 Dec 2014

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
ISSN (Print)1550-4786


ConferenceIEEE International Conference on Data Mining (14th : 2014)


  • GMM
  • permutation
  • top-t ordering
  • Topic model
  • topic segmentation


Dive into the research topics of 'Topic models with topic ordering regularities for topic segmentation'. Together they form a unique fingerprint.

Cite this