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

• Wong, S.K.M.; Butz, C.J., Probabilistic reasoning in a distributed multi-agent environment, Multi Agent Systems, 1998. Proceedings. International Conference on , Jan. 1998 , pp.: 341 -348.*
• Geng, H.; Xiang, Y., Implementation of Fully Distributed Inference in Multiagent MSBN Systems, Electrical and Computer Engineering, 1999 IEEE Canadian Conference on vol.: 3 , Jan. 1999 , pp.: 1698-1703.*
• Luttrell, S.P., An adaptive Bayesian network for texture modelling Texture analysis in radar and sonar, IEE Seminar on , Jan. 1993, pp.: 6/1 -610.*
• Luttrell, S.P.; An adaptive Bayesian network for low-level image processing Artificial Neural Networks, 1993., Third International Conference on , Jan. 1993 , pp.:61 -65.*
• Luttrell, S.P., Partitioned mixture distribution: an adaptive Bayesian network for low-energy image processing, Vision, Image and Signal Processing, IEE Proceedings- vol.: 141 4 , Aug. 1994 , pp.: 251-260.*
• Taniguchi, M.; Haft, M.; Hollmen, J.; Tresp, V., Fraud detection in communication networks using neural and probabilistic methods, Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on, vol.: 2, Jan. 1998 , P.*
• Yamanishi, Kenji, Distributed cooperative Bayesian learning strategies, Proceedings of the tenth annual conference on Computational learning theory , Jan. 1997, pp. 250-262.*
• 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.*
• Heckerman, David, Probabilistic Similarity Networks, MIT Press, Cambridge, Massachusetts, 1990, pp. 53-103.
• Jeffrey D. Banfield and Adrian E. Raferty, Model-Based Gaussian and Non-Gaussian Clustering, pp. 803-821, Sep. (1993).
• P. Cheeseman and J. Stutz, Bavesian Classification AutoClass: Theory and Results, pp. 153-180, AAAI Press (1995).
• David Maxwell Chickering and David Heckerman, Efficient Approximations for the Marginal Likelihood of Bayesian Networks With Hidden Variables, pp. 1-33, Netherlands (1997).
• Nir Friedman, Learning Belief Networks in the Presence of Missing Values and Hidden Variables, Morgan Kaufman (1997).