Industry: text mining with self-organizing maps

ABSTRACT

Today's information age may be characterized by constant massive production and dissemination of written information. More powerful tools for exploring, searching, and organizing the available mass of information are needed to cope with this situation. This need is our starting point for applying data mining techniques on unstructured information as present in text archives. The users will particularly benefit from cluster techniques that uncover similar documents and bring these similarities to the user's attention. In our approach to text mining we suggest relying on the utilization of self-organizing maps for the analysis of a document archive. The benefit of this approach is the intuitive visualization of document similarities thanks to the spatial ordering of the documents within the self-organizing map. We augment the basic capabilities of the neural network with a data description technique that, based on the features learned by the map, automatically selects the most descriptive features of the input patterns mapped onto a particular unit of the map, thus making the associations between the various clusters within the map explicit. We demonstrate the benefits of this approach by using a real-world document archive comprised of articles from Time magazine.

References

1. Agrawal, R. and R. Srikant. 1995. "Mining sequential patterns." In Proceedings of the International Conference on Data Engineering. Taipei, Taiwan. Los Alamitos, CA: IEEE CS Press. pp. 3-14.]]
2. Ahonen, H., O. Heinonen, M. Klemettinen, and A. I. Verkamo. 1997. "Mining the phrasal frontier." In Proceedings of the European Symposium on Principles of Data Mining and Knowledge Discovery. Berlin: Springer-Verlag. pp. 343-350.]]
3. Feldman, R. and I. Dagan. 1995. "Knowledge discovery in textual databases (KDT)." In Proceedings of the International Conference on Knowledge Discovery and Data Mining, Montreal. Menlo Park, CA: AAAI Press. pp. 112-117.]]
4. Feldman, R. and H. Hirsh. 1996. "Mining associations in text in the presence of background knowledge." In Proceedings of the International Conference on Knowledge Discovery and Data Mining, Portland, OR. Menlo Park, CA: AAAI Press. pp. 112-117.]]
5. Feldman, R., W. Klösgen, Y. Ban-Yehuda, G. Kedar, and V. Reznikov. 1997. "Pattern based browsing in document collections." In Proceedings of the European Symposium on Principles of Data Mining and Knowledge Discovery. Berlin: Springer-Verlag. pp. 112-122.]]
6. Hearst, M. A. 1999. "Untangling text data mining." In Proceedings of the Annual Meeting of the Association for Computational Linguistics. College Park, MD. pp. 3-10.]]
7. Honkela, T., S. Kaski, K. Lagus, and T. Kohonen. 1997. "WEBSOM--self-organizing maps of document collections." In Proceedings of the Workshop on Self-Organizing Maps. Espoo, Finland. pp. 298-303.]]
8. Kohonen, T. 1982. "Self-organized formation of topologically correct feature maps." Biol. Cybernet.43: 59-69.]]
9. Kohonen, T. 1995. In Self-organizing Maps. Berlin: Springer-Verlag.]]
10. Kohonen, T. 1998. "Self-organization of very large document collections: state of the art." In Proceedings of the International Conference on Artificial Neural Networks, Skövde, Sweden. Berlin: Springer-Verlag. pp. 65-74.]]
11. Lagus, K., T. Honkela, S. Kaski, and T. Kohonen. 1996. "Self-organizing maps of document collections: a new approach to interactive exploration." In Proceedings of the International Conference on Knowledge Discovery and Data Mining, Portland, OR. Menlo Park, CA: AAAI Press. pp. 238-243.]]
12. Lent, B., R. Agrawal, and R. Srikant. 1997. "Discovering trends in text databases." In Proceedings of the International Conference on Knowledge Discovery and Data Mining. Menlo Park, CA: AAAI Press. pp. 227-230.]]
13. Lin, X., D. Soergel, and G. Marchionini. 1991. "A self-organizing semantic map for information retrieval." In Proceedings of the ACM SIGIR International Conference on R&D in Information Retrieval, Chicago. New York: ACM Press. pp. 262-269.]]
14. Merkl, D. 1995. "A connectionist view on document classification." In Proceedings of the Australasian Database Conference17(2): 153-161.]]
15. Merkl, D. 1997a. "Exploration of document collections with self-organizing maps: a novel approach to similarity representation." In Proceedings of the European Symposium on Principles of Data Mining and Knowledge Discovery, Trondheim, Norway. Berlin: Springer-Verlag. pp. 101-111.]]
16. Merkl, D. 1997b. "Exploration of text collections with hierarchical feature maps." In Proceedings of the International ACM SIGIR Conf on R&D in Information Retrieval, Philadelphia. New York: ACM Press. pp. 186-195.]]
17. Merkl, D. 1998. "Text classification with self-organizing maps: some lessons learned." Neurocomput.21(1-3): 61-77.]]
18. Merkl, D. and A. Rauber. 1999a. "Self-organization of distributed document archives." In Proceedings of the International Database Engineering and Applications Symposium, Montreal, Canada. Los Alamitos, CA: IEEE CS Press. pp. 128-136.]]
19. Merkl, D. and A. Rauber. 1999b. "Uncovering associations between documents." In Proceedings of the IJCAI'99 Workshop on Text Mining. Stockholm. pp. 89-98.]]
20. Merkl, D. and A. Rauber. 2000. "Uncovering the hierarchical structure of text archives by using an unsupervised neural network with adaptive architecture." In Proceedings of the Pacific Asia Conference on Knowledge Discovery and Data Mining, Kyoto, Japan. Berlin: Springer-Verlag. pp. 384-395.]]
21. Rauber, A. and D. Merkl. 1999. "Automatic labeling of self-organizing maps: making a treasure-map reveal its secrets." In Proceedings of the Pacific Asia Conference on Knowledge Discovery and Data Mining, Beijing, China. Berlin: Springer-Verlag. pp. 228-237.]]
22. Roussinov, D. and M. Ramsey. 1998. "Information forage through adaptive visualization." In Proceedings of the ACM International Conference on Digital Libraries, Pittsburgh. New York: ACM Press. pp. 303-304.]]
23. Salton, G. 1989. Automatic Text Processing: The Transformation, Analysis, and Retrieval of Information by Computer. Reading, MA: Addison-Wesley.]]
24. Salton, G. and C. Buckley. 1988. "Term weighting approaches in automatic text retrieval." Inform. Process. Mgmnt24(5): 513-523.]]
25. Swanson, D. R. and N. R. Smalheiser. 1997. "An interactive system for finding complementary literatures: a stimulus to scientific discovery." Artificial Intell.91: 183-203.]]
26. Turtle, H. R. and W. B. Croft. 1992. "A comparison of text retrieval models." Comput. J.35(3): 297-290.]]
27. Baeza-Yates. R. and B. Ribeiro-Neto. 1999. Modern Information Retrieval. Harlow, UK: Addison-Wesley. This book provides an in-depth coverage of contemporary research issues in information retrieval.]]
28. Kohonen, T. 1997. Self-Organizing Maps, 2d ed. Berlin: Springer-Verlag. This is by far the most comprehensive review of self-organizing maps and their applications.]]
29. Ripley, B. 1996. Pattern Recognition and Neural Networks. Cambridge, UK: Cambridge University Press. An excellent introduction to neural networks, which covers a wide spectrum of models and relates them to traditional pattern recognition methods.]]
30. Sparck Jones, K. and P. Willett, 1997. Readings in Information Retrieval. San Francisco, CA: Morgan Kaufmann. A prime source for classic articles in the area of information retrieval.]]

Index Terms

• Information systems

  • Information retrieval

    • Retrieval tasks and goals

      • Clustering and classification

  • Information systems applications

    • Data mining

      • Clustering

    • Multimedia information systems

      • Multimedia databases