Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes

One of the most challenging tasks when adopting Bayesian networks (BNs) is the one of learning their structure from data. This task is complicated by the huge search space of possible solutions and by the fact that the problem is NP-hard. Hence, a full enumeration of all the possible solutions is no...

Full description

Saved in:
Bibliographic Details
Main Authors: Stefano Beretta, Mauro Castelli, Ivo Gonçalves, Roberto Henriques, Daniele Ramazzotti
Format: Article
Language:English
Published: Wiley 2018-01-01
Series:Complexity
Online Access:http://dx.doi.org/10.1155/2018/1591878
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849693110608592896
author Stefano Beretta
Mauro Castelli
Ivo Gonçalves
Roberto Henriques
Daniele Ramazzotti
author_facet Stefano Beretta
Mauro Castelli
Ivo Gonçalves
Roberto Henriques
Daniele Ramazzotti
author_sort Stefano Beretta
collection DOAJ
description One of the most challenging tasks when adopting Bayesian networks (BNs) is the one of learning their structure from data. This task is complicated by the huge search space of possible solutions and by the fact that the problem is NP-hard. Hence, a full enumeration of all the possible solutions is not always feasible and approximations are often required. However, to the best of our knowledge, a quantitative analysis of the performance and characteristics of the different heuristics to solve this problem has never been done before. For this reason, in this work, we provide a detailed comparison of many different state-of-the-art methods for structural learning on simulated data considering both BNs with discrete and continuous variables and with different rates of noise in the data. In particular, we investigate the performance of different widespread scores and algorithmic approaches proposed for the inference and the statistical pitfalls within them.
format Article
id doaj-art-905c74e00f82497aa26fa83e3d5baf6f
institution DOAJ
issn 1076-2787
1099-0526
language English
publishDate 2018-01-01
publisher Wiley
record_format Article
series Complexity
spelling doaj-art-905c74e00f82497aa26fa83e3d5baf6f2025-08-20T03:20:31ZengWileyComplexity1076-27871099-05262018-01-01201810.1155/2018/15918781591878Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic SchemesStefano Beretta0Mauro Castelli1Ivo Gonçalves2Roberto Henriques3Daniele Ramazzotti4DISCo, Universitá degli Studi di Milano-Bicocca, 20126 Milano, ItalyNOVA Information Management School (NOVA IMS), Universidade Nova de Lisboa, Campus de Campolide, 1070-312 Lisboa, PortugalINESC Coimbra, DEEC, University of Coimbra, Coimbra, PortugalNOVA Information Management School (NOVA IMS), Universidade Nova de Lisboa, Campus de Campolide, 1070-312 Lisboa, PortugalDepartment of Pathology, Stanford University, Stanford, California, USAOne of the most challenging tasks when adopting Bayesian networks (BNs) is the one of learning their structure from data. This task is complicated by the huge search space of possible solutions and by the fact that the problem is NP-hard. Hence, a full enumeration of all the possible solutions is not always feasible and approximations are often required. However, to the best of our knowledge, a quantitative analysis of the performance and characteristics of the different heuristics to solve this problem has never been done before. For this reason, in this work, we provide a detailed comparison of many different state-of-the-art methods for structural learning on simulated data considering both BNs with discrete and continuous variables and with different rates of noise in the data. In particular, we investigate the performance of different widespread scores and algorithmic approaches proposed for the inference and the statistical pitfalls within them.http://dx.doi.org/10.1155/2018/1591878
spellingShingle Stefano Beretta
Mauro Castelli
Ivo Gonçalves
Roberto Henriques
Daniele Ramazzotti
Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
Complexity
title Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
title_full Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
title_fullStr Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
title_full_unstemmed Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
title_short Learning the Structure of Bayesian Networks: A Quantitative Assessment of the Effect of Different Algorithmic Schemes
title_sort learning the structure of bayesian networks a quantitative assessment of the effect of different algorithmic schemes
url http://dx.doi.org/10.1155/2018/1591878
work_keys_str_mv AT stefanoberetta learningthestructureofbayesiannetworksaquantitativeassessmentoftheeffectofdifferentalgorithmicschemes
AT maurocastelli learningthestructureofbayesiannetworksaquantitativeassessmentoftheeffectofdifferentalgorithmicschemes
AT ivogoncalves learningthestructureofbayesiannetworksaquantitativeassessmentoftheeffectofdifferentalgorithmicschemes
AT robertohenriques learningthestructureofbayesiannetworksaquantitativeassessmentoftheeffectofdifferentalgorithmicschemes
AT danieleramazzotti learningthestructureofbayesiannetworksaquantitativeassessmentoftheeffectofdifferentalgorithmicschemes