Acquiring Word Similarities with Higher Order Association Mining

  • Sutanu Chakraborti
  • Nirmalie Wiratunga
  • Robert Lothian
  • Stuart Watt
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4626)

Abstract

We present a novel approach to mine word similarity in Textual Case Based Reasoning. We exploit indirect associations of words, in addition to direct ones for estimating their similarity. If word A co-occurs with word B, we say A and B share a first order association between them. If A co-occurs with B in some documents, and B with C in some others, then A and C are said to share a second order co-occurrence via B. Higher orders of co-occurrence may similarly be defined. In this paper we present algorithms for mining higher order co-occurrences. A weighted linear model is used to combine the contribution of these higher orders into a word similarity model. Our experimental results demonstrate significant improvements compared to similarity models based on first order co-occurrences alone. Our approach also outperforms state-of-the-art techniques like SVM and LSI in classification tasks of varying complexity.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Lenz, M., Burkhard, H.: Case Retrieval Nets: Foundations, Properties, Implementation, and Results, Technical Report, Humboldt-Universität zu Berlin (1996)Google Scholar
  2. 2.
    Wiratunga, N., Lothian, R., Chakraborti, S., Koychev, I.: A Propositional Approach to Textual Case Indexing. In: Jorge, A.M., Torgo, L., Brazdil, P.B., Camacho, R., Gama, J. (eds.) PKDD 2005. LNCS (LNAI), vol. 3721, pp. 380–391. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  3. 3.
    Jarmasz, M., Szpakowicz, S.: Roget’s thesaurus and semantic similarity. In: Proceedings of the International Conference on Recent Advances in NLP (RANLP-2003), pp. 212–219 (2003)Google Scholar
  4. 4.
    Lund, K., Burgess, C.: Producing high-dimensional semantic spaces from lexical co-occurrence. Behavior Research, Methods, Instruments and Computers 28(2), 203–208 (1996)Google Scholar
  5. 5.
    Lemaire, B., Denhière, G.: Effects of High-Order Co-occurrences on Word Semantic Similarity. Current Psychology Letters 18(1) (2006)Google Scholar
  6. 6.
    Kontostathis, A., Pottenger, W.M.: A framework for understanding LSI performance. Information Processing and Management 42(1), 56–73 (2006)CrossRefGoogle Scholar
  7. 7.
    Mill, W., Kontostathis, A.: Analysis of the values in the LSI term-term matrix, Technical report, Ursinus College (2004)Google Scholar
  8. 8.
    Mitchell, T.: Machine Learning. Mc Graw Hill International (1997)Google Scholar
  9. 9.
    Porter, M.F.: An algorithm for suffix stripping. Program 14(3), 130–137 (1980)Google Scholar
  10. 10.
    Joachims, T.: Text Categorization with Support Vector Machines: Learning with Many Relevant Features. In: Proc. of ECML, pp. 137–142. ACM Press, New York (1998)Google Scholar
  11. 11.
    Chakraborti, S., Mukras, R., Lothian, R., Wiratunga, N., Watt, S., Harper, D.: Supervised Latent Semantic Indexing using Adaptive Sprinkling. In: Proc. of IJCAI, pp. 1582–1587 (2007)Google Scholar
  12. 12.
    Lamontagne, L.: Textual CBR Authoring using Case Cohesion, in TCBR’06 - Reasoning with Text. In: Proceedings of the ECCBR 2006 Workshops, pp. 33–43 (2006)Google Scholar
  13. 13.
    Edmonds, P.: Choosing the word most typical in context using a lexical co-occurrence network. Meeting of the Association for Computational Linguistics, 507–509 (1997)Google Scholar
  14. 14.
    Lenz, M.: Knowledge Sources for Textual CBR Applications. In: Lenz, M. (ed.) Textual CBR: Papers from the 1998 Workshop Technical Report WS-98-12, pp. 24–29. AAAI Press, Stanford (1998)Google Scholar
  15. 15.
    Semeraro, G., Lops, P., Degemmis, M.: WordNet-based User Profiles for Neighborhood Formation in Hybrid Recommender Systems. In: Procs. of Fifth HIS Conference, pp. 291–296 (2005)Google Scholar
  16. 16.
    Xue, G.-R., Lin, C., Yang, Q., Xi, W., Zeng, H.-J., Yu, Y., Chen, Z.: Scalable Collaborative Filtering Using Cluster-based Smoothing. In: Procs. of the 28th ACM SIGIR Conference, pp. 114–121 (2005)Google Scholar
  17. 17.
    Terra, E., Clarke, C.L.A.: Frequency Estimates for Word Similarity Measures. In: Proceedings of HLT-NAACL 2003, Main Papers, pp. 165–172 (2003)Google Scholar
  18. 18.
    Lenz, M., Burkhard, H.-D.: CBR for Document Retrieval - The FAllQ Project. In: Leake, D.B., Plaza, E. (eds.) Case-Based Reasoning Research and Development. LNCS, vol. 1266, pp. 84–93. Springer, Heidelberg (1997)Google Scholar
  19. 19.
    Budanitsky, A.: Lexical semantic relatedness and its application in natural language processing, Technical Report CSRG390, University of Toronto (1999)Google Scholar
  20. 20.
    Patterson, D., Rooney, N., Dobrynin, V., Galushka, M.: Sophia: A novel approach for textual case-based reasoning. In: Proc. of IJCAI, pp. 1146–1153 (2005)Google Scholar
  21. 21.
    Chakraborti, S., Lothian, R., Wiratunga, N., Watt, S.: Exploiting Higher Order Word Associations in Textual CBR, Technical Report, The Robert Gordon University (2007)Google Scholar
  22. 22.
    Wiratunga, N., Massie, S., Lothian, R.: Unsupervised Textual Feature Selection. In: Roth-Berghofer, T.R., Göker, M.H., Güvenir, H.A. (eds.) ECCBR 2006. LNCS (LNAI), vol. 4106, pp. 340–354. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  23. 23.
    Mori, J., Ishizuka, M., Matsuo, Y.: Extracting Keyphrases To Represent Relations in Social Networks from Web. In: Proc. of the Twentieth IJCAI Conference, pp. 2820–2825 (2007)Google Scholar
  24. 24.
    Boucher-Ryan, P., Bridge, D.: Collaborative Recommending using Formal Concept Analysis. Knowledge-Based Systems 19(5), 309–315 (2006)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Sutanu Chakraborti
    • 1
  • Nirmalie Wiratunga
    • 1
  • Robert Lothian
    • 1
  • Stuart Watt
    • 1
  1. 1.School of Computing, The Robert Gordon University, Aberdeen AB25 1HG, ScotlandUK

Personalised recommendations