Graphical Structure of Bayesian Networks by Eliciting Mental Models of Experts

  • Udai Kumar KudikyalaEmail author
  • Mounika Bugudapu
  • Manasa Jakkula
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 77)


In knowledge-driven approaches to construct Bayesian networks (BN), a knowledge engineer consults with a domain expert to elicit and represent the graphical structure of a BN. The directed graph along with the node probabilities are then used for prediction or diagnosis. In this paper, we present a formal approach to learning the graphical structure of BN using domain expert(s). The proposed PFNetBN technique elicits and represents the mental model of an expert as a directed Pathfinder network. This technique uses the Target method to capture causal/influence relationships among the probability nodes from experts. It then generates a directed graph by applying the Pathfinder algorithm. Consensus Pathfinder network may be generated if multiple experts are involved. This technique generated graphs that are similar to some academic examples from the BN literature. This technique may save time in eliciting and constructing the graphical structure of a BNfrom experts.


Bayesian network Pathfinder network Mental model Directed graph Domain expert Target method 


  1. 1.
    Fenton, N.E., Neil, M.: The use of Bayes and causal modelling in decision making, uncertainty and risk. J. Council Eur. Prof. Inf. Soc. (CEPIS) 12(5), 10–21 (2011)Google Scholar
  2. 2.
    Nadkarni, S., Prakash, P.P.: A causal mapping approach to constructing Bayesian networks. Decis. Support Syst. 38(2), 259–281 (2004)CrossRefGoogle Scholar
  3. 3.
    Yet, B., Bastani, K., Raharjo, H., Lifvergren, S., William, Marsh W., Bergman, B.: Decision support system for Warfarin therapy management using Bayesian networks. Decis. Support Syst. 55, 488–498 (2013)CrossRefGoogle Scholar
  4. 4.
    Fenton, N.E., Neil, M.: Decision support software for probabilistic risk assessment using Bayesian networks. IEEE Softw. 31(2), 21–26 (2014)Google Scholar
  5. 5.
    Neil, M., Fenton, N.E., Nielson, L.: Buliding large-scale bayesian networks. Knowl. Eng. Rev. 15(3), 257–284 (2000)CrossRefGoogle Scholar
  6. 6.
    Helsper, E.M., Van der Gaag, L.C., Feelders, A.J., Loeffen, W.L.A., Geenen, P.L., Elbers, A.R.W.: Bringing order into Bayesian-network construction. In: The Proceedings of the 3rd International Conference on Knowledge Capture, K-CAP’05, Banff, Alberta, Canada, pp. 121–128, Oct 2–5 (2005)Google Scholar
  7. 7.
    Velikovaa, M., Lucasa, P.J.F., Samulskic, M., Karssemeijerd, N.: On the interplay of machine learning and background knowledge in image interpretation by Bayesian networks. Artif. Intell. Med. 57, 73–86 (2013)CrossRefGoogle Scholar
  8. 8.
    Yet, B., Perkins, Z., Fenton, N., Tai, N., Marsh, W.: Not just data: a method for improving prediction with knowledge. J. Biomed. Inf. 48, 28–37 (2014)CrossRefGoogle Scholar
  9. 9.
    Constantinou, A.C., Fenton, N., Marsh, W., Radlinski, L.: From complex questionnaire and interviewing data to intelligent Bayesian network models for medical decision support. Artif. Intell. Med. 67, 75–93 (2016)CrossRefGoogle Scholar
  10. 10.
    Heckerman, D.E., Nathwani, B.N.: Toward normative expert systems: part II probability based representations for efficient knowledge acquisition and inference. Methods Inf. Med. 31, 106–116 (1992)CrossRefGoogle Scholar
  11. 11.
    Xio-xuan, H., Hui, W., Shuo, W.: Using expert’s knowledge to build bayesian network. In: IEEE International Conference on Computational Intelligence and Security Workshops, pp. 220–223 (2007)Google Scholar
  12. 12.
    Henrion, M.: Practical issues constructing a Bayes’ belief network. In: The Proceeding of Uncertainty in AI ‘87, Seattle, WA, pp. 132–140 (1987)Google Scholar
  13. 13.
    Tossell, C.C., Schvaneveldt, R.W., Branaghan, R.J.: Targeting knowledge structures: a new method to elicit the relatedness of concepts. Cogn. Technol. 1(2), 11–19 (2010)Google Scholar
  14. 14.
    Dearholt, D.W., Schvaneveldt, R.W.: Properties of pathfnder networks. In: Schvaneveldt, R.W. (ed.) Pathfinder Associative Networks: Studies in Knowledge Organization, Ablex Publishing Corp. Norwood, NJ, USA. pp. 1–30 (1990)Google Scholar
  15. 15.
    Rubin, D.C.: Directed graphs as memory representations: the case of rhyme. In: Schvaneveldt, R.W. (ed.) Pathfinder associative networks: studies in knowledge organization, pp. 1–30. Ablex Publishing Corp, Norwood (1990)Google Scholar
  16. 16.
    Kudikyala, U.K., Vaughn, R.B.: Software requirements understanding pathfinder networks: discovering and evaluating mental models. J. Syst. Softw. 74(1), 101–108 (2005)CrossRefGoogle Scholar
  17. 17.
    Dearholt, D.W., Alt, K.J., Halpin, R.F., Oliver, R.L.: Foundational aspects of student-controlled learning: a paradigm for design, development, and assessment appropriate for web-based instruction. J. Eng. Edu. 93, 129–138 (2004)Google Scholar
  18. 18.

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Udai Kumar Kudikyala
    • 1
    Email author
  • Mounika Bugudapu
    • 1
  • Manasa Jakkula
    • 1
  1. 1.Computer Science and Engineering DepartmentKeshav Memorial Institute of TechnologyNarayanguda, HyderabadIndia

Personalised recommendations