An Algorithm to Define the Node Probability Functions of Bayesian Networks based on Ranked Nodes
Joa˜o Nunes, Renan Willamy, Mirko Perkusich, Renata Saraiva, Kyller Gorgonio, Hyggo Almeida, Angelo Perkusich "An Algorithm to Define the Node Probability Functions of Bayesian Networks based on Ranked Nodes", International Journal of Engineering Trends and Technology (IJETT), V52(3),151-156 October 2017. ISSN:2231-5381. www.ijettjournal.org. published by seventh sense research group
Bayesian Network (BN) has been used in a broad range of applications. A challenge in constructing a BN is defining the node probability tables (NPTs), which can be learned from data or elicited from domain experts. In practice, it is common to not have enough data for learning and elicitation from experts is the only option. However, the complexity of defining NPTs grows exponentially, making their elicitation process costly and error-prone. Previous work proposed a solution: the ranked nodes method (RNM). However, the details necessary to implement it were not presented. Nowadays, the solution is only available through a commercial tool. Hence, this paper presents an algorithm to define NPT using the RNM. We include details regarding sampling and how to mix truncated Normal distributions and convert the resulting distribution into an NPT. We compared the results calculated using our algorithm with the commercial tool through an experiment. The results show that our solution is equivalent to the commercial tools’ in terms of NPT definition with a mean difference of 1.6%. Furthermore, our solution is faster. The solution developed is made available as open source software.
 I. Ben-Gal. Bayesian Networks. John Wiley and Sons, 2007.  A. Constantinou and N. Fenton. Towards smart-data: Improving predictive accuracy in long-term football team performance. Knowledge-Based Systems, pages –, 2017.
 A. C. Constantinou, N. E. Fenton, and M. Neil. Profiting from an inefficient association football gambling market: Prediction, risk and uncertainty using bayesian networks. Knowledge-Based Systems, 50:60 – 86, 2013.
 B. Das. Generating Conditional Probabilities for Bayesian Networks: Easing the Knowledge Acquisition Problem. Computing Research Repository, cs.AI/0411034, 2004.
 F. J. Diez. Parameter adjustment in bayes networks. the generalized noisy or-gate. In Proceedings of the 9th Conference on Uncertainty in Artificial Intelligence, pages 99–105. Morgan Kaufmann, 1993.
 N. Fenton and M. Neil. Risk Assessment and Decision Analysis with Bayesian Networks. CRC Press, 5 edition, 11 2012.
 N. E. Fenton, M. Neil, and J. G. Caballero. Using ranked nodes to model qualitative judgments in bayesian networks. IEEE Trans. on Knowl. and Data Eng., 19(10):1420–1432, Oct. 2007.
 N. Friedman, D. Geiger, and M. Goldszmidt. Bayesian network classifiers. Machine Learning, 29(2-3):131– 163, 1997.
 D. Heckerman. Learning in graphical models. chapter A Tutorial on Learning with Bayesian Networks, pages 301–354. MIT Press, Cambridge, MA, USA, 1999.
 K. Huang and M. Henrion. Efficient search-based inference for noisyor belief networks: Topepsilon. In Proceedings of the Twelfth International Conference on Uncertainty in Artificial Intelligence, UAI’96, pages 325–331, San Francisco, CA, USA, 1996. Morgan Kaufmann Publishers Inc.
 E. Lee, Y. Park, and J. G. Shin. Large engineering project risk management using a bayesian belief network. Expert Syst. Appl., 36(3):5880–5887, Apr. 2009.
 J. Pearl and S. Russell. Bayesian networks. Handbook of brain theory and neural networks, 1995.
 M. Perkusich, K. Gorgonio, H. Almeida, and A. Perkusich. Assisting the continuous improvement of scrum projects using metrics and bayesian networks. Journal of Software: Evolution and Process, 2016. Article in Press.
 M. Perkusich, G. Soares, H. Almeida, and A. Perkusich. A procedure to detect problems of processes in software development projects using bayesian networks. Expert Systems with Applications, 42(1):437 – 450, 2015.
Bayesian Network; Expert systems; Node Probability Table; Ranked nodes.