Skip to main content
Log in

Explaining inferences in Bayesian networks

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

While Bayesian network (BN) can achieve accurate predictions even with erroneous or incomplete evidence, explaining the inferences remains a challenge. Existing approaches fall short because they do not exploit variable interactions and cannot account for compensations during inferences. This paper proposes the Explaining BN Inferences (EBI) procedure for explaining how variables interact to reach conclusions. EBI explains the value of a target node in terms of the influential nodes in the target’s Markov blanket under specific contexts, where the Markov nodes include the target’s parents, children, and the children’s other parents. Working back from the target node, EBI shows the derivation of each intermediate variable, and finally explains how missing and erroneous evidence values are compensated. We validated EBI on a variety of problem domains, including mushroom classification, water purification and web page recommendation. The experiments show that EBI generates high quality, concise and comprehensible explanations for BN inferences, in particular the underlying compensation mechanism that enables BN to outperform alternative prediction systems, such as decision tree.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Pearl J (1988) Probabilistic reasoning in intelligent systems: networks of plausible inference. Kaufmann, San Mateo

    Google Scholar 

  2. Yap G-E, Tan A-H, Pang H-H (2006) Discovering causal dependencies in mobile context-aware recommenders. In: Proceedings of MDM’06 (in CD-ROM Procs), Nara, Japan, May 2006. IEEE Computer Society, Los Alamitos

    Google Scholar 

  3. Teach RL, Shortliffe EH (1981) An analysis of physician attitudes regarding computer-based clinical consultation systems. Comput Biomed Res 14:542–558

    Article  Google Scholar 

  4. Suermondt HJ (1992) Explanation in Bayesian belief networks. PhD thesis, Medical Information Sciences, Stanford University, Palo Alto, CA, March 1992

  5. Haddawy P, Jacobson J, Kahn C Jr (1997) BANTER: A Bayesian network tutoring shell. Artif Intell Med 10:177–200

    Article  Google Scholar 

  6. Humphreys P (1989) The chances of explanation: causal explanation in the social, medical, and physical sciences. Princeton University Press, Princeton

    Google Scholar 

  7. Strevens M (2006) Scientific explanation. Macmillan encyclopedia of philosophy, 2nd edn

  8. Wong SKM, Ziarko W (1986) INFER—An adaptive decision support system based on the probabilistic approximate classification. In: Proceedings of the 6th international workshop on expert systems and their applications, Avignon, pp 713–725

  9. Boutilier C, Friedman N, Goldszmidt M, Koller D (1996) Context-specific independence in Bayesian networks. In: Proceedings of UAI’96, Reed College, Portland, OR, USA, August 1996. Kaufmann, Los Altos, pp 115–123

    Google Scholar 

  10. Quinlan JR (1993) C4.5: programs for machine learning. Kaufmann, Los Altos

    Google Scholar 

  11. Tversky A, Kahneman D (1974) Judgement under uncertainty: heuristics and biases. Read Uncertain Reason, pp 32–39

  12. Pennington N, Hastie R (1988) Explanation-based decision making: effects of memory structure on judgment. JEP: Learn Mem Cogn 14(3):521–533

    Article  Google Scholar 

  13. Newman DJ, Hettich S, Blake CL, Merz CJ (1998) UCI repository of machine learning databases. Irvine: University of California, Department of Information and Computer Sciences. URL: http://www.ics.uci.edu/~mlearn/MLRepository.html

  14. Pazzani MJ, Muramatsu J, Billsus D (1996) Syskill & Webert: Identifying interesting web sites. In: Proceedings of AAAI’96, IAAI’96, Portland, OR, pp 54–61

  15. Jensen FV, Kjærulff U, Olesen KG, Pedersen J (1989) An expert system for control of waste water treatment—a pilot project. Technical report, Judex Datasystemer A/S, Aalborg, Denmark (in Danish)

  16. Zhang NL, Poole D (1999) On the role of context-specific independence in probabilistic inference. In: Proceedings of IJCAI’99. Stockholm, Sweden, pp 1288–1293

  17. Butz CJ (2002) Exploiting contextual independencies in web search and user profiling. In: Proceedings of WCCI’02, Honolulu, HI, USA, pp 1051–1056

  18. Poole D, Zhang NL (2003) Exploiting contextual independence in probabilistic inference. J Artif Intell Res 18:263–313

    MATH  MathSciNet  Google Scholar 

  19. Korb KB, Nicholson AE (2003) Bayesian artificial intelligence. CRC Press, Boca Raton

    Google Scholar 

  20. Lacave C, Diez FJ (2002) A review of explanation methods for Bayesian networks. Knowl Eng Rev 17:107–127

    Article  Google Scholar 

  21. Chajewska U, Draper DL (1998) Explaining predictions in Bayesian networks and influence diagrams. In: AAAI spring symposium, Stanford Univ., Palo Alto, CA

  22. Zhou Z, Liu H, Li SZ, Chua CS (2001) Rule mining with prior knowledge: a belief networks approach. Intell Data Anal 5(2):95–110

    MATH  Google Scholar 

  23. Druzdzel M, Henrion M (1990) Using scenarios to explain probabilistic inference. In: Working notes of AAAI’90 workshop on explanation, pp 133–141

  24. Koiter JR (2006) Visualizing inference in Bayesian networks. PhD thesis, Delft University of Technology, Delft, The Netherlands

  25. Shachter RD (1986) Evaluating influence diagrams. Oper Res 34(6):871–882

    Article  MathSciNet  Google Scholar 

  26. Olmsted SM (1983) On representing and solving decision problems. PhD thesis, Dept. of Engineering-Economic Systems, Stanford University, Palo Alto, CA

  27. Zhang NL (1998) Inference in Bayesian networks: the role of context-specific independence. Technical Report HKUST-CS98-09, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

  28. Ibarguengoytia P, Sucar L, Vadera S (1996) A probabilistic model for sensor validation. In: Proceedings of UAI’96, Reed College, USA. Kaufmann, Los Altos, pp 332–339

    Google Scholar 

  29. Wallace CS, Korb KB (1999) Learning linear causal models by MML sampling. In: Causal models and intelligent data management. Springer, Berlin

    Google Scholar 

  30. Norsys Software Corporation (2005) Netica-Java application programmer interfaces (Netica-J API). URL: http://www.norsys.com/netica-j.html

  31. Salton G (1971) The SMART retrieval system—experiments in automatic document processing. Prentice Hall, Englewood Cliffs

    Google Scholar 

  32. Pazzani MJ, Billsus D (1997) Learning and revising user profiles: the identification of interesting web sites. Mach Learn 27:313–331

    Article  Google Scholar 

  33. Weiss SM, Indurkhya N (1991) Reduced complexity rule induction. In: Proceedings of IJCAI’91, Sydney, Australia, August 1991. Kaufmann, Los Altos, pp 678–684

    Google Scholar 

  34. Towell GG, Shavlik JW (1992) Interpretation of artificial neural networks: mapping knowledge-based neural networks into rules. Adv NIPS 4:977–984

    Google Scholar 

  35. Towell GG, Shavlik JW (1993) Extracting refined rules from knowledge-based neural networks. Mach Learn 13:71–101

    Google Scholar 

  36. Zhang NL, Poole D (1996) Exploiting causal independence in Bayesian network inference. J Artif Intell Res 5:301–328

    MATH  MathSciNet  Google Scholar 

  37. Moret BME (1982) Decision trees and diagrams. Comput Surv 14(4):593–623

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ghim-Eng Yap.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Yap, GE., Tan, AH. & Pang, HH. Explaining inferences in Bayesian networks. Appl Intell 29, 263–278 (2008). https://doi.org/10.1007/s10489-007-0093-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-007-0093-8

Keywords

Navigation