Abstract
In real applications of health data management, it is necessary to make Bayesian network (BN) inferences when evidence is not contained in existing conditional probability tables (CPTs). In this paper, we are to augment the learning function to BN inferences from existing CPTs. Based on the support vector machine (SVM) and sigmoid, we first transform existing CPTs into samples. Then we use transformed samples to train the SVM for finding a maximum likelihood hypothesis, and to fit a sigmoid for mapping outputs of the SVM into probabilities. Further, we give the approximate inference method of BNs with maximum likelihood hypotheses. An applied example and preliminary experiments show the feasibility of our proposed methods.
This work was supported by the National Natural Science Foundation of China (No. 60763007), the Chun-Hui Project of the Educational Department of China (No. Z2005-2-65003), the Natural Science Foundation of Yunnan Province (No. 2005F0009Q), and the Cultivation Project for Backbone Teachers of Yunnan University.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Pearl, J.: Probabilistic reasoning in intelligent systems: network of plausible inference. Morgen Kaufmann publishers, Inc., San Mates, California (1998)
Liu, W.Y., Yue, K., Zhang, J.D.: Augmenting Learning Function to Bayesian Network Inferences with Maximum Likelihood Parameters. Technical Report, Yunnan University (2007)
Mitchell, T.M.: Machine Learning. McGraw-Hill Companies, Inc., New York (1997)
Burges, C.J.C.: A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery 2(2), 121–167 (1998)
Chang, C., Liu, C.J.: Training v-support vector classifiers: theory and algorithms. Neural Computing 13(9), 2119–2147 (2001)
Platt, J.C.: Probabilistic Outputs for Support Vector Machines and Comparisons to Regularized Likelihood Methods. In: Advances in Large Margin Classifiers. MIT Press, Cambridge (1999)
Heckerman, D., Wellman, M.P.: Bayesian networks. Communication of the ACM 38(3), 27–30 (1995)
Russel, S.J., Norving, P.: Artificial intelligence - a modern approach. Pearson Education Inc., Publishing as Prentice-Hall (2002)
Cheng, J., Greiner, R., Kelly, J., Bell, D., Liu, W.: Learning Bayesian network from data: an information-theory based approach. Artificial Intelligence 137(2), 43–90 (2002)
Buntine, W.L.: A guide to the literature on learning probabilistic networks from data. IEEE Trans. on Knowl. Data Eng. 8(2), 195–210 (1996)
Heckerman, D., Mandani, A., Wellman, M.P.: Real-world applications of Bayesian networks. Communications of the ACM 38(3), 25–30 (1995)
Vapnik, V.: Statistical learning theory. John Wiley and Sons, Inc., New York (1998)
Ross, S.M.: Simulation, 3rd edn. Academic Press, Inc., London (2002)
Asuncion, A., Newman, D.J.: UCI Machine Learning Repository. Irvine, CA: University of California, Department of Information and Computer Science (2007), http://www.ics.uci.edu/~mlearn/MLRepository.html
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhang, J., Yue, K., Liu, W. (2008). Learning-Function-Augmented Inferences of Causalities Implied in Health Data. In: Ishikawa, Y., et al. Advanced Web and Network Technologies, and Applications. APWeb 2008. Lecture Notes in Computer Science, vol 4977. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-89376-9_9
Download citation
DOI: https://doi.org/10.1007/978-3-540-89376-9_9
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-89375-2
Online ISBN: 978-3-540-89376-9
eBook Packages: Computer ScienceComputer Science (R0)