Abstract
A dynamic Bayesian network (DBN) is one of popular approaches for relational knowledge discovery such as modeling relations or dependencies, which change over time, between variables of a dynamic system. In this paper, we propose an adaptive learning method (autoDBN) to learn DBNs with changing structures from multivariate time series. In autoDBN, segmentation of time series is achieved first through detecting geometric structures transformed from time series, and then model regions are found from the segmentation by designed finding strategies; in each found model region, a DBN model is established by existing structure learning methods; finally, model revisiting is developed to refine model regions and improve DBN models. These techniques provide a special mechanism to find accurate model regions and discover a sequence of DBNs with changing structures, which are adaptive to changing relations between multivariate time series. Experimental results on simulated and real time series show that autoDBN is very effective in finding accurate/reasonable model regions and gives lower error rates, outperforming the switching linear dynamic system method and moving window method.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Murphy KP (2002) Dynamic Bayesian networks: representation, inference and learning. PhD thesis, University of California Berkeley. http://www.cs.ubc.ca/~murphyk
Russell S, Norvig P (2003) Artificial intelligence: a modern approach, 2nd edn. Pearson Education, Inc., Upper Saddle River
Schuster A, Wolff R, Trock D (2005) A high-performance distributed algorithm for mining association rules. Knowl Inf Syst 7(4): 458–475
Berti-Equille L (2007) Data quality awareness: a case study for cost optimal association rule mining. Knowl Inf Syst 11(2): 191–215
Pearl J (1988) Probabilistic reasoning in intelligent systems. Morgan Kauffman, San Mateo
Neapolitan RE (2003) Learning Bayesian networks. Prentice Hall, Englewood Cliffs
Chen R, Sivakumar K, Kargupta H (2004) Collective mining of Bayesian networks from distributed heterogeneous data. Knowl Inf Syst 6(2): 164–187
Pena JM, BjÄorkegren J, Tegner J (2005) Learning dynamic Bayesian network models via cross- validation. Pattern Recognit Lett 26(14): 2295–2308
Yu J, Smith VA, Wang PP, Hartemink AJ, Jarvis ED (2004) Advances to Bayesian network inference for generating causal networks from observational biological data. Bioinformatics 20(18): 3594–3603
Ledin J (2001) Simulation engineering. CMP Books, The Netherlands
Tucker A, Liu X (2004) A Bayesian network approach to explaining time series with changing structure. Intell Data Anal 8(5): 469–480
Pavlovic V, Rehg JM, Cham TJ, Murphy KP (1999) A dynamic Bayesian network approach to figure tracking using learned dynamic models. In: ICCV 1999, pp 94–101
Pavlovic V, Rehg JM, MacCormick J (2000) Learning switching linear models of human motion. In: NIPS 2000, pp 981–987
Barber D (2006) Expectation correction for smoothed inference in switching linear dynamical systems. J Mach Learn Res 7: 2515–2540
Chen W (2001) An introduction to differential manifold. High Education Press, Beijing
Mo G, Liu K (2003) Methodology of function approximation. Science Press, Beijing
Mei X, Huang M (2003) Differential geometry. Beijing Normal University Press, Beijing
Walpole RE, Myers RH, Myers SL, Ye K (2002) Probability and statistics for engineers and scientists, 7th edn. Pearson Education, Inc., Upper Saddle River
Murphy KP (2006). Bayes net toolbox for MATLAB. http://bnt.sourceforge.net/
Leray P, Francois O (2004) BNT structure learning package. http://bnt.insa-rouen.fr/ajouts.html
Maharaj E (2002) A pattern recognition of time series using wavelets. In: 15th Computational Statistics Conference of the International Association of Statistical Computing, Berlin
StatLib repository: http://lib.stat.cmu.edu/ http://www.liacc.up.pt/~ltorgo/Regression/stock.tgz
Jin R, Goswami A, Agrawal G (2007) Fast and exact out-of-core and distributed k-means clustering. Knowl Inf Syst 10(1): 17–40
Author information
Authors and Affiliations
Corresponding author
Additional information
An erratum to this article can be found at http://dx.doi.org/10.1007/s10115-008-0175-x
Rights and permissions
About this article
Cite this article
Wang, K., Zhang, J., Shen, F. et al. Adaptive learning of dynamic Bayesian networks with changing structures by detecting geometric structures of time series. Knowl Inf Syst 17, 121–133 (2008). https://doi.org/10.1007/s10115-008-0124-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10115-008-0124-8