Learning Local Components to Understand Large Bayesian Networks

Yifeng Zeng, Yanping Xiang, Jorge Cordero H., Yujian Lin

    Research output: Contribution to conferencePaper

    119 Downloads (Pure)

    Abstract

    Bayesian networks are known for providing an intuitive and compact representation of probabilistic information and allowing the creation of models over a large and complex domain. Bayesian learning and reasoning are nontrivial for a large Bayesian network. In parallel, it is a tough job for users (domain experts) to extract accurate information from a large Bayesian network due to dimensional difficulty. We define a formulation of local components and propose a clustering algorithm to learn such local components given complete data. The algorithm groups together most interrelevant attributes in a domain. We evaluate its performance on three benchmark Bayesian networks and provide results in support. We further show that the learned components may represent local knowledge more precisely in comparison to the full Bayesian networks when working with a small amount of data.
    Original languageEnglish
    Pages1076-1081
    Number of pages6
    DOIs
    Publication statusPublished - Dec 2009
    Event9th IEEE International Conference on Data Mining (ICDM) - Miami Beach, United States
    Duration: 6 Dec 20099 Dec 2009
    Conference number: 9th

    Conference

    Conference9th IEEE International Conference on Data Mining (ICDM)
    CountryUnited States
    Period6/12/099/12/09

    Fingerprint Dive into the research topics of 'Learning Local Components to Understand Large Bayesian Networks'. Together they form a unique fingerprint.

  • Cite this

    Zeng, Y., Xiang, Y., H., J. C., & Lin, Y. (2009). Learning Local Components to Understand Large Bayesian Networks. 1076-1081. Paper presented at 9th IEEE International Conference on Data Mining (ICDM), United States. https://doi.org/10.1109/ICDM.2009.120