Research on Optimization of Markov Blanket Algorithm Based on Bayesian Network
DOI:
https://doi.org/10.63313/JCSFT.9011Keywords:
Markov blanket, IPC-MB algorithm, IAMB, algorithm, Feature selection, Algorithm optimizationAbstract
With the advent of the big data era, the scale and complexity of data are con-stantly increasing, and feature selection has become one of the key issues in improving the performance and computational efficiency of machine learning models.15 The Markov carpet algorithm is widely used in feature selection and data dimensionality reduction tasks because it can effectively describe the con-ditional dependencies among variables in Bayesian networks. This paper pro-poses an optimization method for the computational efficiency and accuracy problems of the existing IPC-MB and IAMB algorithms in high-dimensional data processing. The optimization methods include the optimization of sorting strat-egies, the improvement of filtering strategies, and the application of symmetry principles. Through theoretical analysis and experimental verification, this pa-per proves the performance improvement of the optimized algorithm on multi-ple datasets. The optimized IPC-MB and IAMB algorithms are significantly supe-rior to the original algorithms in terms of computing time and accuracy. The experimental results show that the optimized algorithm can efficiently process high-dimensional data and has strong stability and application prospects.
References
[1] Friedman,N.,Geiger,D.,&Goldszmidt,M.(1997).Bayesian network classifiers.Machine Learning,29(2):131-163.
[2] Chickering,D.M.(2002).Learning Bayesian networks is NP-complete.Proceedings of the 13th Conference on Uncertainty in Artificial Intelligence(pp.150-157).
[3] Heckerman,D.,Geiger,D.,&Chickering,D.M.(1995).Learning Bayesian networks:The combi-nation of knowledge and statistical data.Machine Learning,20(3):197-243.
[4] Koller,D.,&Friedman,N.(2009).Probabilistic graphical models:Principles and tech-niques.MIT Press.
[5] Tian,J.,&Pearl,J.(2002).On the testable implications of local independence in graphical models.Journal of Machine Learning Research,3:125-147.
[6] Heckerman,D.,&Meek,C.(2004).Causal inference using Bayesian networks.Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence(pp.385-392).
[7] Larranaga,P.,&Husten,R.(2013).Machine learning in probabilistic graphical models:An overview.Journal of Machine Learning Research,14(1):1-14.
[8] Silva,J.P.,&Shenoy,P.P.(2009).Fast algorithms for computing Markov blankets.Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence(pp.230-239).
[9] Buntine,W.(1996).A guide to the literature on learning probabilistic networks from da-ta.IEEE Transactions on Knowledge and Data Engineering,8(1):195-210.
[10] Foulds,J.,&Koller,D.(2010).A study of Markov blanket algorithms for gene expression da-ta.Bioinformatics,26(13):1634-1642.
[11] Zhang,X.,&Wang,Y.(2014).Efficient Markov blanket discovery for large-scale da-tasets.Machine Learning,94(3):535-563.
[12] Geman,D.,&Geman,S.(1984).Stochastic relaxation,Gibbs distributions,and the Bayesian restoration of images.IEEE Transactions on Pattern Analysis and Machine Intelli-gence,6(6):721-741.
[13] Koller,D.,&Sahami,M.(1996).Toward optimal feature selection.Proceedings of the 13th In-ternational Conference on Machine Learning(pp.284-292).
[14] Shibata,T.,&Konishi,S.(2012).Feature selection based on Markov blanket discovery.Journal of Machine Learning Research,13:1473-1492.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 by author(s) and Erytis Publishing Limited.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
