Applying a Burst Model to Detect Bursty Topics in a Topic Model

  • Yusuke Takahashi
  • Takehito Utsuro
  • Masaharu Yoshioka
  • Noriko Kando
  • Tomohiro Fukuhara
  • Hiroshi Nakagawa
  • Yoji Kiyota
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7614)

Abstract

This paper focuses on two types of modeling of information flow in news stream, namely, burst analysis and topic modeling. First, when one wants to detect a kind of topics that are paid much more attention than usual, it is usually necessary for him/her to carefully watch every article in news stream at every moment. In such a situation, it is well known in the field of time series analysis that Kleinberg’s modeling of bursts is quite effective in detecting burst of keywords. Second, topic models such as LDA (latent Dirichlet allocation) are also quite effective in estimating distribution of topics over a document collection such as articles in news stream. However, Kleinberg’s modeling of bursts is usually applied only to bursts of keywords but not to those of topics. Considering this fact, we propose how to apply Kleinberg’s modeling of bursts to topics estimated by a topic model such as LDA and DTM (dynamic topic model).

Keywords

Time Series News Topic Model Kleinberg’s Burst Model 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kleinberg, J.: Bursty and hierarchical structure in streams. In: Proc. 8th SIGKDD, pp. 91–101 (2002)Google Scholar
  2. 2.
    Blei, D.M., Ng, A.Y., Jordan, M.I.: Latent Dirichlet allocation. Journal of Machine Learning Research 3, 993–1022 (2003)MATHGoogle Scholar
  3. 3.
    Blei, D.M., Lafferty, J.D.: Dynamic topic models. In: Proc. 23rd ICML, pp. 113–120 (2006)Google Scholar
  4. 4.
    Mane, K., Borner, K.: Mapping topics and topic bursts in PNAS. Proc. PNAS. 101(suppl. 1), 5287–5290 (2004)CrossRefGoogle Scholar
  5. 5.
    AlSumait, L., Barbará, D., Gentle, J., Domeniconi, C.: Topic Significance Ranking of LDA Generative Models. In: Buntine, W., Grobelnik, M., Mladenić, D., Shawe-Taylor, J. (eds.) ECML PKDD 2009, Part I. LNCS, vol. 5781, pp. 67–82. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  6. 6.
    Wang, X., Zhai, C., Hu, R.S.: Mining correlated bursty topic patterns from coordinated text streams. In: Proc. 13th SIGKDD, pp. 784–793 (2007)Google Scholar
  7. 7.
    AlSumait, L., Bardara, D., Domeniconi, C.: On-Line LDA: Adaptive topic models for mining text streams with applications to topic detection and tracking. In: Proc. 8th ICDM, pp. 3–12 (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Yusuke Takahashi
    • 1
  • Takehito Utsuro
    • 1
  • Masaharu Yoshioka
    • 2
  • Noriko Kando
    • 3
  • Tomohiro Fukuhara
    • 4
  • Hiroshi Nakagawa
    • 5
  • Yoji Kiyota
    • 6
  1. 1.University of TsukubaTsukubaJapan
  2. 2.Hokkaido UniversitySapporoJapan
  3. 3.National Institute of InformaticsTokyoJapan
  4. 4.National Institute of Advanced Industrial Science and TechnologyTokyoJapan
  5. 5.University of TokyoTokyoJapan
  6. 6.NEXT Co., Ltd.TokyoJapan

Personalised recommendations