Abstract
Recent years, data-intensive computing has become a research hotspot. It also proposed a new challenge to traditional Bayesian inference methods. It is known that, traditional Bayesian inference methods could do a good job when all data stay in a single station. However, when come to data-intensive computing, it would hard to apply to new situation directly because data often been distributed in multi stations under data intensive computing. Among different Bayesian inference methods, random algorithm often been regarded as a common and effective one. And the sampling method adopted in random algorithm would largely influence the efficiency of this random algorithm. Gibbs sampling method often been used in random algorithm for Bayesian inference. Taking all of this into consideration, an improved Bayesian inference method for data-intensive computing is developed in this paper, which first use improved Gibbs sampling method in each station to gain the suitable information, then union them together to infer the final result. The validity of this method is discussed in theory and illustrated by experiment.
This paper is sponsored by: The National Natural Science Foundation of China(No.61163003), the Yunnan Provincial Department of Education Foundation(No.2011Y029).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Perkins, A.: Data intensive computing. In: Proceedings of the 2006 ACM/IEEE Conference on Supercomputing (2006)
Chen, K., Jiang, W., Martin: A note on some algorithms for the Gibbs posterior. Statistics and Probability Letters 80, 1234–1241 (2010)
van Hasselt, M.: Bayesian inference in a sample selection model. Journal of Econometrics, 221–232 (2011)
Omori, Y.: Efficient Gibbs sampler for Bayesian analysis of a sample selection model. Statistics and Probability Letters, 1300–1311 (2007)
Kottegoda, N.T., Natale, L., Raiteri, E.: Gibbs sampling of climatic trends and periodicities. Journal of Hydrology, 54–64 (2007)
Eastern forum of science and technology. What is data-intensive computing (2010), http://www.efst.sh.cn/showKnowledge.do?id=577
Agrawal, D., El Abbadi, A., Antony, S., Das, S.: Data Management Challenges in Cloud Computing Infrastructures. In: Kikuchi, S., Sachdeva, S., Bhalla, S. (eds.) DNIS 2010. LNCS, vol. 5999, pp. 1–10. Springer, Heidelberg (2010)
Pearl: Probabilistic reasoning in intelligent systems: networks of plausible inference, pp. 116–131. Morgen Daufmann Publishers Inc., California (1998)
Russell, S.: Artificial IntelligenceA modern method, 2nd edn. The People’s Posts and Telecommunications Press (2010)
Liu, W., Li, W., Yue, K.: Intelligent data analysis. Science Press, Beijing (2007)
Deshpande, S.: Probabilistic graphical models and their role in database. In: Proc. of VLDB 2007, pp. 1435–1436 (2007)
Tuncozgur, B., Elbeyli, L., Gungor, A., Isik, F., Akay, H.: Chest wall reconstruction with autologas rib grafts in dogs and report of a clinic case. European Journal of Cardio-Thoracic Surgery 16(3), 292–295 (1999)
Pericchi, L.R.: Model Selection and Hypothesis Testing based on Objective Probabilities and Bayes Factors. In: Handbook of Statistics, vol. 25, pp. 115–149 (2005)
Tsukuma, H.: Generalized Bayes minimax estimation of the normal mean matrix with unknown covariance matrix. Journal of Multivariate Analysis, 2296–2304 (2009)
Chen, L., Yang, M.: Empirical Bayes testing for equivalence. Journal of Statistical Planning and Inference, 2670–2681 (2011)
Liu, W., Han, C., Shi, Y.: Unsupervised Learning for Finite Mixture Models Via Modi ed Gibbs Sampling. Journal of Xi’an Jiaotong University, 15–19 (2009)
Lauritzen, S.L., Spiegelhalter, D.J.: Local computations with probabilities on graphical structures and their application to expert systems. Royal Statistics Society B, 157–194 (1988)
Zhang, L., Guo, H.: Introduction to Bayesian Networks. Science Press, Beijing (2006)
Lauritzen, S.: Local computation with probabilities on graphical structures and their application to expert systems. Roy. Stat. B, 157–224 (1988)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ma, F., Liu, W. (2012). An Improved Bayesian Inference Method for Data-Intensive Computing. In: Li, Z., Li, X., Liu, Y., Cai, Z. (eds) Computational Intelligence and Intelligent Systems. ISICA 2012. Communications in Computer and Information Science, vol 316. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-34289-9_16
Download citation
DOI: https://doi.org/10.1007/978-3-642-34289-9_16
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-34288-2
Online ISBN: 978-3-642-34289-9
eBook Packages: Computer ScienceComputer Science (R0)