An example-based mapping method for text categorization and retrieval

ABSTRACT

A unified model for text categorization and text retrieval is introduced. We use a training set of manually categorized documents to learn word-category associations, and use these associations to predict the categories of arbitrary documents. Similarly, we use a training set of queries and their related documents to obtain empirical associations between query words and indexing terms of documents, and use these associations to predict the related documents of arbitrary queries. A Linear Least Squares Fit (LLSF) technique is employed to estimate the likelihood of these associations. Document collections from the MEDLINE database and Mayo patient records are used for studies on the effectiveness of our approach, and on how much the effectiveness depends on the choices of training data, indexing language, word-weighting scheme, and morphological canonicalization. Alternative methods are also tested on these data collections for comparison. It is evident that the LLSF approach uses the relevance information effectively within human decisions of categorization and retrieval, and achieves a semantic mapping of free texts to their representations in an indexing language. Such a semantic mapping lead to a significant improvement in categorization and retrieval, compared to alternative approaches.

REFERENCES

• 1: CHUTE, C. G., AND YANG, Y. 1992. An evaluatmn of concept based Latent Semantic Indexing for clinical information retrieval. In Proceedings of the 16th Annual Symposzum on Computer Applications ~n Medical Care, vol. 16. McGraw-HilL New York, 639-643.
• 2: CPHA. 1986. International Classifice, tion of Dzseases. 9th Rev. Clinical Modifications. Commission on Professional and Hospital Activities. Ann Arbor, Mich.
• 3: DEERWESTER, S., DUMAIS, S. T., FURNAS, G.W., LANDAUER, T. K., AND HARSHMAN, R. 1990. Indexing by Latent Semantic analysis. J. Am. Soc. Inf. Sci. 41, 6, 391-407.
• 4: DONGARRA, J. J., MOLER, C. B., BUNCH, J. R., AND STEWART, C.W. 1979. LINPACK Users' Guide. SIAM, Philadelphia, Pa.
• 5: DSC. 1991. M++ Class Library, User Guide. Rel. 3. Dyad Software Corporation, Bellevue, Wash.
• 6: EVANS, D. A., CHUTE, C. G., HANDERSON, S. K., YANG, Y., MONARCH, I. A., AND HERSH, W. R. 1992. Mapping vocabularies using "Latent Semantics." In MEDINFO 92. 1462-1468.
• 7: EVANS, D. A., HERSH, W. R., MONARCH, I. A., LEFFERTS, R. G., AND HANDERSON, S.K. 1991. Automatic indexing of abstracts via natural-language processing using a simple thesaurus. Medical Decision Making 11, 4, 108-115.
• 8: Norbert Fuhr , Chris Buckley, A probabilistic learning approach for document indexing, ACM Transactions on Information Systems (TOIS), v.9 n.3, p.223-248, July 1991 [doi>10.1145/125187.125189]
• 9: FUHR, N., ET AL. 1991. AIR/X--a rule-based multistage indexing systems for large subject fields. In Proceedings of the RIAO'91. 606-623.
• 10: Gene H. Golub , Charles F. Van Loan, Matrix computations (3rd ed.), Johns Hopkins University Press, Baltimore, MD, 1996
• 11: HAYNES, R., McKSBBON, K., WALKER, C., RYAN, N., FITZGERALD, D., AND RAMSDEN, M. 1990. Online access to MEDLINE in clinical settings. Ann. Int. Med. 112, 1, 78 84.
• 12: HERSH, W. R., HICKAM, D. H., AND LEONE, T.J. 1992. Words, concepts, or both: Optimal indexing units for automated information retrieval. In Proceedings of the 16th Annual Symposium on Computer AppDcations in Medical Core, voL 16. McGraw-Hill, New York, 644 648.
• 13: LAWSON, C. L., AND HANSON, R. J. 1974. Solving Least Squares Problems. Prentice-Hall, Englewood Cliffs~ N.J.
• 14: David D. Lewis, Evaluating text categorization, Proceedings of the workshop on Speech and Natural Language, p.312-318, February 19-22, 1991, Pacific Grove, California [doi>10.3115/112405.112471]
• 15: NLM. 1993. Medical Subject Headings (MESH). National Library of Medicine, Bethesda, Md.
• 16: SALTON, G. 1991. Development in automatic text retrieval. Science 253, 974-980.
• 17: Gerard Salton, Automatic text processing: the transformation, analysis, and retrieval of information by computer, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1989
• 18: SALTON, G., AND BUCKLEY, C. 1990. Improving retrieval performance by relevance feedback. J. Am. Soc. Inf. Sci. 41, 4, 288-297.
• 19: Yiming Yang , Christopher G. Chute, An application of least squares fit mapping to text information retrieval, Proceedings of the 16th annual international ACM SIGIR conference on Research and development in information retrieval, p.281-290, June 27-July 01, 1993, Pittsburgh, Pennsylvania, United States [doi>10.1145/160688.160738]
• 20: YANG, Y., AND CHUTE, C.G. 1993b. Words or concepts: The features of indexing units and their optimal use in information retrieval. In Proceedings of the 17th Annual Symposium on Computer Apphcations tn Medical Cure, vol. 17. McGraw-Hill, New York, 685-689.
• 21: Yiming Yang , Christopher G. Chute, A Linear Least Squares Fit mapping method for information retrieval from natural language texts, Proceedings of the 14th conference on Computational linguistics, August 23-28, 1992, Nantes, France [doi>10.3115/992133.992139]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.1 Content Analysis and Indexing
          Subjects: Indexing methods

Additional Classification:
  G. Mathematics of Computing
  G.1 NUMERICAL ANALYSIS
      G.1.2 Approximation
          Subjects: Least squares approximation

  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval
          Subjects: Retrieval models

  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.6 Learning
          Subjects: Parameter learning

General Terms:
Experimentation, Theory

Keywords:
document categorization, query categorization, statistical learning of human decisions