Abstract

One aspect of the invention is the construction of mixtures of Bayesian networks. Another aspect of the invention is the use of such mixtures of Bayesian networks to perform inferencing. A mixture of Bayesian networks (MBN) consists of plural hypothesis-specific Bayesian networks (HSBNs) having possibly hidden and observed variables. A common external hidden variable is associated with the MBN, but is not included in any of the HSBNs. The number of HSBNs in the MBN corresponds to the number of states of the common external hidden variable, and each HSBN is based upon the hypothesis that the common external hidden variable is in a corresponding one of those states. In one mode of the invention, the MBN having the highest MBN score is selected for use in performing inferencing. In another mode of the invention, some or all of the MBNs are retained as a collection of MBNs which perform inferencing in parallel, their outputs being weighted in accordance with the corresponding MBN scores and the MBN collection output being the weighted sum of all the MBN outputs. In one application of the invention, collaborative filtering may be performed by defining the observed variables to be choices made among a sample of users and the hidden variables to be the preferences of those users.

Other References

• Larranaga, P.; Poza, M.; Yurramendi, Y.; Murga, R.H.; Kuijpers, C.M.H., Structure learning of Bayesian networks by genetic algorithms: a performance analysis of control parameters, Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol.: 1, Sep. 1996.
• Banfield, Jeffrey D., and Raferty, Adrian E., "Model-Based Gaussian and Non-Gaussian Clustering," Biometrics, vol. 49, Sep. 1993, pp. 803-821
• Cheeseman, P., and Stutz, J., "Bayesian Classification (AutoClass): Theory and Results," AAAI Press, 1995, pp. 153-180
• Chickering, David Maxwell and Heckerman, David, "Efficient Approximations for the Marginal Likelihood of Bayesian Networks with Hidden Variables," Machine Learning, vol. 1, Kluwer Academic Publishers, Boston, 1997, pp. 1-33
• Friedman, Nir, "Learning Belief Networks in the Presence of Missing Values and Hidden Variables," Proceedings of the 14th Annual Conference on Machine Learning, Morgan Kauffman, San Francisco, CA, 1997
• Heckerman, David E., "Probabilistic Similarity Networks," MIT Press, Cambridge, Massachusetts, 1990, pp. 53-103
• Jordan, Michael I., A statistical approach to decision tree modeling, Annual Workshop on Computational Learning Theory, Proceedings of the seventh annual ACM conference on Computational learning theory, Jul. 12-15, 1994, New Brunswick, NJ USA, pp. 13.
• Laskey, K.B., Sensitivity analysis for probability assessments in Bayesian networks, Systems, Man and Cybernectics, IEEE Transactions on, vol.: 25 6, Jun. 1995, pp.: 901-909.
• Goodman, G.L.; McMichael, D.W., A Gaussian Mixture Model Classifier Using Supervised And Unsupervised Learning, Signal Processing and Its Applications, 1996. ISSPA 96., Fourth International Symposium on, vol.: 2, pp.: 565-566, Jan. 1996.
• McMichael, D.W., Estimating Gaussian Mixture Models from Data with Missing Features, Signal Processing and Its Applications, 1996. ISSPA 96., Fourth International Symposium on, vol.: 1, pp.: 377-378, Jan. 1996.
• McMichael, D.W., Bayesian growing and pruning strategies for MAP-optimal estimation of Gaussian mixture models, Artificial Neural Networks, 1995., Fourth International Conference on, 1995, pp.: 364-368, Jan. 1995.
• Laskey, K.B., A Bayesian approach to clustering and classification, Systems, Man, and Cybernetics, 1991. Decision Aiding for Complex Systems, Conference Proceedings., 1991 IEEE International Conference on, 1991, pp.: 179-183 vol. 1, Jan. 1991.
• Laskey, K.B., Cohen, M.S.; Martin, A.W., Representing and eliciting knowledge about uncertain evidence and its implications, Systems, Man and Cybernetics, IEEE Transactions on, vol.: 19 3, May-Jun. 1989, pp.: 536-54