Textual Energy of Associative Memories: Performant Applications of Enertex Algorithm in Text Summarization and Topic Segmentation

  • Silvia Fernández
  • Eric SanJuan
  • Juan Manuel Torres-Moreno
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4827)


In this paper we present a Neural Network approach, inspired by statistical physics of magnetic systems, to study fundamental problems of Natural Language Processing (NLP). The algorithm models documents as neural network whose Textual Energy is studied. We obtained good results on the application of this method to automatic summarization and Topic Segmentation.


Automatic Summarization Topic Segmentation Statistical Methods Statistical Physics 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hopfield, J.: Neural networks and physical systems with emergent collective computational abilities. National Academy of Sciences 9, 2554–2558 (1982)CrossRefMathSciNetGoogle Scholar
  2. 2.
    Hertz, J., Krogh, A., Palmer, G.: Introduction to the theorie of Neural Computation. Addison Wesley, Redwood City, CA (1991)Google Scholar
  3. 3.
    Salton, G., McGill, M.: Introduction to modern information retrieval. Computer Science Series. McGraw-Hill, New York (1983)MATHGoogle Scholar
  4. 4.
    Ma, S.: Statistical Mechanics. World Scientific, Philadelphia, CA (1985)MATHGoogle Scholar
  5. 5.
    Kosko, B.: Bidirectional associative memories. IEEE Transactions Systems Man, Cybernetics 18, 49–60 (1988)CrossRefMathSciNetGoogle Scholar
  6. 6.
    Porter, M.: An algorithm for suffix stripping. Program 14, 130–137 (1980)Google Scholar
  7. 7.
    Manning, C.D., Schutze, H.: Foundations of Statistical Natural Language Processing. The MIT Press, Cambridge (2000)Google Scholar
  8. 8.
    Mani, I., Maybury, M.T.: Automatic Text Summarization. MIT Press, Cambridge (1999)Google Scholar
  9. 9.
    Radev, D., Winkel, A., Topper, M.: Multi Document Centroid-based Text Summarization. In: ACL 2002 (2002)Google Scholar
  10. 10.
    Torres-Moreno, J.M., Velázquez-Morales, P., Meunier, J.: Condensés de textes par des méthodes numériques. In: JADT. vol. 2, pp. 723–734 (2002)Google Scholar
  11. 11.
    Lin, C.Y.: Rouge: A package for automatic evaluation of summaries. In: WAS 2004 (2004)Google Scholar
  12. 12.
    Amini, M.R., Zaragoza, H., Gallinari, P.: Learning for sequence extraction tasks. In: RIAO 2000 pp. 476–489 (2000)Google Scholar
  13. 13.
    Caillet, M., Pessiot, J.F., Amini, M., Gallinari, P.: Unsupervised learning with term clustering for thematic segmentation of texts. In: RIAO 2004, pp. 648–657 (2004)Google Scholar
  14. 14.
    Chuang, S.L., Chien, L.F.: A practical web-based approach to generating Topic hierarchy for Text segments. In: ACM IKM, Washington, pp. 127–136 (2004)Google Scholar
  15. 15.
    Sitbon, L., Bellot, P.: Segmentation thématique par chaînes lexicales pondérées. In: TALN 2005, vol. 1 pp. 505–510 (2005)Google Scholar
  16. 16.
    Brants, T., Chen, F., Tsochantaridis, I.: Topic-based document segmentation with probabilistic latent semantic anaysis. In: CIKM 2002, Virginia, USA, pp. 211–218 (2002)Google Scholar
  17. 17.
    Galley, M., McKeown, K.R., Fosler-Lussier, E., Jing, H.: Discourse segmentation of multi-party conversation. In: ACL-2003, Sapporo, Japan, pp. 562–569 (2003)Google Scholar
  18. 18.
    Pevzner, L., Hearst, M.: A critique and improvement of an evaluation metric for text segmentation. In: Computational Linguistic. vol. 1, pp. 19–36 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Silvia Fernández
    • 1
    • 2
  • Eric SanJuan
    • 1
  • Juan Manuel Torres-Moreno
    • 1
    • 3
  1. 1.Laboratoire Informatique d’Avignon, BP 1228 F-84911 Avignon Cedex 9France
  2. 2.Laboratoire de Physique des Matériaux, CNRS UMR 7556, NancyFrance
  3. 3.École Polytechnique de Montréal - Département de génie informatique, CP 6079 Succ. Centre Ville H3C 3A7, Montréal (Québec)Canada

Personalised recommendations