Improving Bayesian network local structure learning via data-driven symmetry correction methods

Learning the structure of a Bayesian network (BN) from data is NP-hard. To efficiently handle high-dimensional datasets, many BN local structure learning algorithms are proposed. These learning algorithms can be categorized into two types: constraint-based and score-based. These learning algorithms...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhao, Jianjun, Ho, Shen-Shyang
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/151697
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-151697
record_format dspace
spelling sg-ntu-dr.10356-1516972021-06-29T00:15:14Z Improving Bayesian network local structure learning via data-driven symmetry correction methods Zhao, Jianjun Ho, Shen-Shyang School of Computer Science and Engineering Engineering::Computer science and engineering Bayesian Network Symmetry Correction Learning the structure of a Bayesian network (BN) from data is NP-hard. To efficiently handle high-dimensional datasets, many BN local structure learning algorithms are proposed. These learning algorithms can be categorized into two types: constraint-based and score-based. These learning algorithms learn the local structures separately for each node. As a result, asymmetric pairs of neighbors and Markov blankets create conflicts between the local structures. To resolve the conflicts, symmetry correction is required. The commonly used AND-rule symmetry correction method, which simply drops nodes in asymmetric pairs from the neighbor sets and Markov blankets of both nodes, may result in loss of information in learning the BN. In this paper, we propose a hybrid framework that combines a local structure learning algorithm of a particular type (either constraint-based or score-based) with a data-driven symmetry correction method of the other type. The score-based symG method and the constraint-based symC method are proposed to be used in the hybrid framework. Empirical results show that performances of constraint-based learning algorithms are improved by using the proposed score-based symG method. Similarly, the performance of score-based local learning algorithm is better when symC is used, compared to using symG. This research was supported by a Singapore International Graduate Award. 2021-06-29T00:15:14Z 2021-06-29T00:15:14Z 2019 Journal Article Zhao, J. & Ho, S. (2019). Improving Bayesian network local structure learning via data-driven symmetry correction methods. International Journal of Approximate Reasoning, 107, 101-121. https://dx.doi.org/10.1016/j.ijar.2019.02.004 0888-613X https://hdl.handle.net/10356/151697 10.1016/j.ijar.2019.02.004 2-s2.0-85061808849 107 101 121 en International Journal of Approximate Reasoning © 2019 Elsevier Inc. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Bayesian Network
Symmetry Correction
spellingShingle Engineering::Computer science and engineering
Bayesian Network
Symmetry Correction
Zhao, Jianjun
Ho, Shen-Shyang
Improving Bayesian network local structure learning via data-driven symmetry correction methods
description Learning the structure of a Bayesian network (BN) from data is NP-hard. To efficiently handle high-dimensional datasets, many BN local structure learning algorithms are proposed. These learning algorithms can be categorized into two types: constraint-based and score-based. These learning algorithms learn the local structures separately for each node. As a result, asymmetric pairs of neighbors and Markov blankets create conflicts between the local structures. To resolve the conflicts, symmetry correction is required. The commonly used AND-rule symmetry correction method, which simply drops nodes in asymmetric pairs from the neighbor sets and Markov blankets of both nodes, may result in loss of information in learning the BN. In this paper, we propose a hybrid framework that combines a local structure learning algorithm of a particular type (either constraint-based or score-based) with a data-driven symmetry correction method of the other type. The score-based symG method and the constraint-based symC method are proposed to be used in the hybrid framework. Empirical results show that performances of constraint-based learning algorithms are improved by using the proposed score-based symG method. Similarly, the performance of score-based local learning algorithm is better when symC is used, compared to using symG.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Zhao, Jianjun
Ho, Shen-Shyang
format Article
author Zhao, Jianjun
Ho, Shen-Shyang
author_sort Zhao, Jianjun
title Improving Bayesian network local structure learning via data-driven symmetry correction methods
title_short Improving Bayesian network local structure learning via data-driven symmetry correction methods
title_full Improving Bayesian network local structure learning via data-driven symmetry correction methods
title_fullStr Improving Bayesian network local structure learning via data-driven symmetry correction methods
title_full_unstemmed Improving Bayesian network local structure learning via data-driven symmetry correction methods
title_sort improving bayesian network local structure learning via data-driven symmetry correction methods
publishDate 2021
url https://hdl.handle.net/10356/151697
_version_ 1703971215997140992