|
Published Articles >> Table of Contents >> Abstract
2003 IEEE/WIC International Conference on Web Intelligence (WI'03)
p. 606
Research on Modeling with Dynamic Bayesian Networks
Fengzhan Tian, Tsinghua University
Hongwei Zhang, Tsinghua University
Yuchang Lu, Tsinghua University
Full Article Text:
  
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/WI.2003.1241278
Send link to a friend
| Abstract |
|
For simplicity of calculation, Dynamic Bayesian Networks (DBNs) make assumptions that their evolvement follows Markov process and the transition probabilities in the evolvement are time-invariant. While this is not the case in many real complex systems. For the purpose of modeling these complex systems with DBNs, the paper attempts to add hidden variables to the evolutional process so as to build Markov models and expands the EM-EA algorithm to the DBNs learning in the presence of hidden variables. Moreover, as for the time-variant transition probabilities, the paper estimates the sufficient statistics of posterior time slices using polynomial fitting algorithm, and then learns the time-variant transition probabilities with both current sufficient statistics and estimated sufficient statistics. The theoretical analysis demonstrates the validity of the methods of the paper. The future work is to do experimental
analysis with real complex systems.
|
 Additional Information
|
Citation:
Fengzhan Tian, Hongwei Zhang, Yuchang Lu,
"Research on Modeling with Dynamic Bayesian Networks,"
wi,
p. 606,
2003 IEEE/WIC International Conference on Web Intelligence (WI'03),
2003
|
|
