Trie methods for representing text

  • T. H. Merrett
  • Heping Shang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 730)


We propose a new trie organization for large text documents requiring secondary storage. Index size is critical in all trie representations of text, and our organization is smaller than all known methods. Access time is as good as the best known method. Tries can be constructed in good time. For an index of 100 million entries, our experiments show size factors of less than 3, as compared with 3.4 for the best previous method. Our measurements show expected access costs of 0.1 sec., and construction times of 18 to 55 hours, depending on the text characteristics.

Our organization can also handle dynamic data, and we give new algorithms for inserting and deleting. It supports searches for general patterns, as well as a variety of special searches, such as proximity, range, longest repetitions and most frequent occurrences.

Key words and phrases

PATRICIA pattern matching secondary storage text trie 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A. V. Aho, R. Sethi, and J. D. Ullman. Compilers Principles, Techniques, and Tools. Addison-Wesley Publishing Co., Reading, MA, 1986.Google Scholar
  2. [2]
    R. de la Briandais. File searching using variable-length keys. In Proc. Western Joint Computer Conf., pages 295–8, San Francisco, March 1959.Google Scholar
  3. [3]
    L. Devroye. A note on the average depth of tries. Computing, 28:367–371, 1982.Google Scholar
  4. [4]
    E. H. Fredkin. Trie memory. Communications of the ACM, 3(9):490–9, Sept. 1960.Google Scholar
  5. [5]
    G. H. Gonnet. Efficient searching of text and pictures. Technical Report OED-88-02, Centre for the New Oxford English Dictionary, University of Waterloo, Waterloo, Ont., Canada, 1988.Google Scholar
  6. [6]
    G. H. Gonnet, R. A. Baeza-Yates, and T. Snider. Lexicograhic indices for text: Inverted files vs. PAT trees. Technical Report OED-91-01, Centre for the New Oxford English Dictionary, University of Waterloo, Waterloo, Ont., Canada, February 1991.Google Scholar
  7. [7]
    D. E. Knuth. The Art of Computer Programming. Addison-Wesley Publishing Co., Reading, Mass., 1968–1973. Volumes I, II, III.Google Scholar
  8. [8]
    T. H. Merrett and H. Shang. Trie methods for representing text. Technical Report TR-SOCS-93.3, McGill University, School of Computer Science, June 1993.Google Scholar
  9. [9]
    D. R. Morrison. PATRICIA: Practical algorithm to retrieve information coded in alphanumeric. Journal of the ACM, 15:514–34, 1968.Google Scholar
  10. [10]
    J. A. Orenstein. Blocking mechanism used by multidimensional tries. Unpublished Letter, February 1983.Google Scholar
  11. [11]
    B. Pittel. Asymptotical growth of a class of random trees. The Annals of Probability, 13(2):414–427, 1985.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • T. H. Merrett
    • 1
  • Heping Shang
    • 1
  1. 1.School of Computer ScienceMcGill UniversityMontréal

Personalised recommendations