Visualizing Discovered Rule Sets with Visual Graphs Based on Compressed Entropy Density

  • Einoshin Suzuki
  • Hiroki Ishihara
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1711)

Abstract

This paper presents a post-processing algorithm of rule discovery for augmenting the readability of a discovered rule set. Rule discovery, in spite of its usefulness as a fundamental data-mining technique, outputs a huge number of rules. Since usefulness of a discovered rule is judged by human inspection, augmenting the readability of a discovered rule set is an important issue. We formalize this problem as a transformation of a rule set into a tree structure called a visual graph. A novel information-based criterion which represents compressed entropy of a data set per description length of the graph is employed in order to evaluate the readability quantitatively. Experiments with an agricultural data set in cooperation with domain experts confirmed the effectiveness of our method in terms of readability and validness.

Keywords

Association Rule Domain Expert Conclusion Node Visual Graph Rule Discovery 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Agrawal, R., Mannila, H., Srikant, R., Toivonen, H., Verkamo, A.I.: Fast Discovery of Association Rules. In: Advances in Knowledge Discovery and Data Mining, pp. 307–328. AAAI Press/MIT Press, Menlo Park/Calif. (1996)Google Scholar
  2. 2.
    Brachman, R.J., Anand, T.: The Process of Knowledge Discovery in Databases. In: Advances in Knowledge Discovery and Data Mining, pp. 37–57. AAAI Press/MIT Press, Menlo Park/Calif. (1996)Google Scholar
  3. 3.
    Dougherty, J., Kohavi, R., Sahami, M.: Supervised and Unsupervised Discretization of Continuous Features. In: Proc. Twelfth Int’l Conf. Machine Learning (ICML 1995), pp. 194–202. Morgan Kaufmann, San Francisco (1995)Google Scholar
  4. 4.
    Fayyad, U.M., Piatetsky-Shapiro, G., Smyth, P.: From Data Mining to Knowledge Discovery: An Overview. In: Advances in Knowledge Discovery and Data Mining, pp. 1–34. AAAI Press/MIT Press, Menlo Park/Calif. (1996)Google Scholar
  5. 5.
    Gaines, B.R.: Transforming Rules and Trees. In: Advances in Knowledge Discovery and Data Mining, pp. 205–226. AAAI Press/MIT Press, Menlo Park/Calif. (1996)Google Scholar
  6. 6.
    Jensen, F.V.: An Introduction to Bayesian Networks. Springer, New York (1996)Google Scholar
  7. 7.
    Smyth, P., Goodman, R.M.: An Information Theoretic Approach to Rule Induction from Databases. IEEE Trans. Knowledge and Data Eng. 4(4), 301–316 (1992)CrossRefGoogle Scholar
  8. 8.
    Suzuki, E.: Autonomous Discovery of Reliable Exception Rules. In: Proc. Third Int’l Conf. Knowledge Discovery & Data Mining (KDD 1997), pp. 259–262. AAAI Press, Menlo Park (1997)Google Scholar
  9. 9.
    Suzuki, E.: Simultaneous Reliability Evaluation of Generality and Accuracy for Rule Discovery in Databases. In: Proc. Fourth Int’l Conf. Knowledge Discovery & Data Mining (KDD 1998), pp. 339–343. AAAI Press, Menlo Park (1998)Google Scholar
  10. 10.
    Suzuki, E., Kodratoff, Y.: Discovery of Surprising Exception Rules based on Intensity of Implication. In: Żytkow, J.M. (ed.) PKDD 1998. LNCS, vol. 1510, pp. 10–18. Springer, Heidelberg (1998)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Einoshin Suzuki
    • 1
  • Hiroki Ishihara
    • 1
  1. 1.Division of Electrical and Computer Engineering, Faculty of EngineeringYokohama National UniversityYokohamaJapan

Personalised recommendations