Machine Learning

, Volume 9, Issue 1, pp 57–94 | Cite as

The Utility of Knowledge in Inductive Learning

  • Michael Pazzani
  • Dennis Kibler


In this paper, we demonstrate how different forms of background knowledge can be integrated with an inductive method for generating function-free Horn clause rules. Furthermore, we evaluate, both theoretically and empirically, the effect that these forms of knowledge have on the cost and accuracy of learning. Lastly, we demonstrate that a hybrid explanation-based and inductive learning method can advantageously use an approximate domain theory, even when this theory is incorrect and incomplete.

Learning relations combining inductive and explanation-based learning 


  1. Bergadano, F., & Giordana, A. (1988). A knowledge intensive approach to concept induction. Proceedingsofhe Fifth International Conference on Machine Learning (pp. 305–317). Ann Arbor, MI: Morgan Kaufmann.Google Scholar
  2. Bergadano, F., Giordana, A., & Ponsero, S. (1989). Deduction in top-down inductive learning. Proceedingsofthe Sixth International Workshop on Machine Learning (pp. 23–25). Ithaca, NY: Morgan Kaufmann.Google Scholar
  3. Brunk, C., & Pazzani, M. (1991). An investigation of noise tolerant relational learning algorithms. Proceedingsof the Eighth International Workshop on Machine Learning (pp. 389–393). Evanston, IL: MorganKaufmann.Google Scholar
  4. Cohen, W. (in press). Abductive explanation-based learning: A solution to the multiple inconsistentexplanationproblem. Machine Learning.Google Scholar
  5. Danyluk, A. (1989). Finding new rules for incomplete theories: Explicit biases for induction with contextualinformation. Proceedings of the Sixth International Workshop on Machine Learning (pp. 34–36).Ithaca, NY: Morgan Kaufmann.Google Scholar
  6. DeJong, G., & Mooney, R. (1986). Explanation-based learning: An alternate view. MachineLearning, 1, 145–176.Google Scholar
  7. Flann, N., & Dietterich, T. (1989). A study of explanation-basd methods for inductive learning. MachineLearning, 4, 187–226.Google Scholar
  8. Ginsberg, M., & Harvey, W. (1990). Iterative broadening. Proceedings of the Eighth NationalConference onArtificial Intelligence (pp. 216–220). Boston, MA: Morgan Kaufmann.Google Scholar
  9. Hirsh, H. (1989). Combining empirical and analytical learning with version spaces. Proceedings of theSixthInternational Workshop on Machine Learning (pp. 29–33). Ithaca, NY: Morgan Kaufmann.Google Scholar
  10. Katz, B. (1989). Inegrating learning in a neural network. Proceedings of the Sixth International WorkshoponMachine Learning (pp. 69–71). Ithaca, NY: Morgan Kaufmann.Google Scholar
  11. Korf, R.E. (1985). Depth-first iterative-deepening: An optimal admissible tree search. ArtificialIntelligence, 1, 11–46.Google Scholar
  12. Lebowitz, M. (1986). Integrated learning: Controlling explanation. Cognitive Science, 10.Google Scholar
  13. Michalski, R. (1980). Pattern recognition as rule-guided inference. IEEE Transactions on PatternAnalysis and Machine Intelligence, 2, 349–361.Google Scholar
  14. Mitchell, T., Keller, R., & Kedar-Cabelli, S. (1986). Explanation-based learning: A unifying view. MachineLearning, 1,47–80.Google Scholar
  15. Mooney, R., & Ourston, D. (1989). Induction over the unexplained: Integrated learning of concepts with bothexplainable and conventional aspects. Proceedings of the Sixth International Workshop on MachineLearning (pp. 5–7). Ithaca, NY: Morgan Kaufmann.Google Scholar
  16. Muggleton, S., Bain, M., Hayes-Michie, J., & Michie, D. (1989). An experimental comparison of humanand machine learning formalisms. Proceedings of the Sixth International Workshop on Machine Learning(pp. 115–118). Ithaca, NY: Morgan Kaufmann.Google Scholar
  17. Muggleton, S., & Feng, C. (1990). Efficient induction of logic programs. Proceedings of the FirstConference on Algorithmic Learning Theory. Tokyo, Japan: OhmshaGoogle Scholar
  18. Ourston, D., & Mooney, R. (1990). Chaining the rules: A comprehensive approach to theory refinement.Proceedings of the Eighth National Conference on Artificial Intelligence (pp. 815–820). Boston,MA: Morgan Kaufmann.Google Scholar
  19. Pagallo, G., & Haussler, D. (1990). Boolean feature discovery in empirical learning. MachineLearning, 5, 71–100.Google Scholar
  20. Pazzani, M. (1989). Explanation-based learning with weak domain theories. Proceedings of the SixthInternational Workshop on Machine Learning (pp. 72–74). Ithaca, NY: Morgan Kaufmann.Google Scholar
  21. Pazzani, M.J. (1990). Creating a memory of causal relationships: An integration of empirical andexplanation-based learning methods. Hillsdale, NJ: Lawrence Erlbaum Associates.Google Scholar
  22. Pazzani, M., & Brunk, C. (1990). Detecting and correcting errors in rule-based expert systems: An integrationof empirical and explanation-based learning. Proceedings of the Workshop on Knowledge Acquisition forKnowledge-Based Systems. Banff, Canada.Google Scholar
  23. Pazzani, M., Brunk, C., & Silverstein, G. (1991). A knowledge-intensive approach to relational conceptlearning. Proceedings of the Eighth International Workshop on Machine Learning (pp. 432–436).Evanston, IL: Morgan Kaufmann.Google Scholar
  24. Quinlan, J.R. (1986). Induction of decision trees. Machine Learning, 1, 81–106.Google Scholar
  25. Quinlan, J.R. (1989). Learning relations: A comparison of a symbolic and a connectionist approach(Technical Report). Sydney, Australia: University of Sydney.Google Scholar
  26. Quinlan, J.R. (1990). Learning logical definitions from relations. Machine Learning, 5,239–266.Google Scholar
  27. Sarrett, W., & Pazzani, M. (1989). One-sided algorithms for integrating empirical and explanation-basedlearning. Proceedings of the Sixth International Workshop on Machine Learning (pp. 26–28).Ithaca, NY: Morgan Kaufmann.Google Scholar
  28. Shavlik, J., & Towell, G. (1989). Combining explanation-based learning and artificial neural networks. Proceedingsof the Sixth International Workshop on Machine Learning (pp. 90–93). Ithaca, NY: MorganKaufmann.Google Scholar
  29. Silverstein, G., & Pazzani, M. (1991). Relational clichés: Constraining constructive induction duringrelational learning. Proceedings of the Eighth International Workshop on Machine Learning (pp. 203–207).Evanston, IL: Morgan Kaufmann.Google Scholar
  30. Widmer, G. (1990). Incremental knowledge-intensive learning: A case study based on an extension toBergadano & Giordana's integrated learning strategy (Technical Report). Austrian Research Institute forArtificial Intelligence.Google Scholar

Copyright information

© Kluwer Academic Publishers 1992

Authors and Affiliations

  • Michael Pazzani
    • 1
  • Dennis Kibler
    • 1
  1. 1.Department of Information & Computer ScienceUniversity of California, IrvineIrvine

Personalised recommendations