Cargando…

Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring

BACKGROUND: The problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These dat...

Descripción completa

Detalles Bibliográficos
Autores principales: Jiang, Xia, Jao, Jeremy, Neapolitan, Richard
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4666609/
https://www.ncbi.nlm.nih.gov/pubmed/26624895
http://dx.doi.org/10.1371/journal.pone.0143247
_version_ 1782403725268090880
author Jiang, Xia
Jao, Jeremy
Neapolitan, Richard
author_facet Jiang, Xia
Jao, Jeremy
Neapolitan, Richard
author_sort Jiang, Xia
collection PubMed
description BACKGROUND: The problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These data present new challenges. In particular, it is difficult to discover predictive variables, when each variable has little marginal effect. An example concerns Genome-wide Association Studies (GWAS) datasets, which involve millions of single nucleotide polymorphism (SNPs), where some of the SNPs interact epistatically to affect disease status. Towards determining these interacting SNPs, researchers developed techniques that addressed this specific problem. However, the problem is more general, and so these techniques are applicable to other problems concerning interactions. A difficulty with many of these techniques is that they do not distinguish whether a learned interaction is actually an interaction or whether it involves several variables with strong marginal effects. METHODOLOGY/FINDINGS: We address this problem using information gain and Bayesian network scoring. First, we identify candidate interactions by determining whether together variables provide more information than they do separately. Then we use Bayesian network scoring to see if a candidate interaction really is a likely model. Our strategy is called MBS-IGain. Using 100 simulated datasets and a real GWAS Alzheimer’s dataset, we investigated the performance of MBS-IGain. CONCLUSIONS/SIGNIFICANCE: When analyzing the simulated datasets, MBS-IGain substantially out-performed nine previous methods at locating interacting predictors, and at identifying interactions exactly. When analyzing the real Alzheimer’s dataset, we obtained new results and results that substantiated previous findings. We conclude that MBS-IGain is highly effective at finding interactions in high-dimensional datasets. This result is significant because we have increasingly abundant high-dimensional data in many domains, and to learn causes and perform prediction/classification using these data, we often must first identify interactions.
format Online
Article
Text
id pubmed-4666609
institution National Center for Biotechnology Information
language English
publishDate 2015
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-46666092015-12-10 Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring Jiang, Xia Jao, Jeremy Neapolitan, Richard PLoS One Research Article BACKGROUND: The problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These data present new challenges. In particular, it is difficult to discover predictive variables, when each variable has little marginal effect. An example concerns Genome-wide Association Studies (GWAS) datasets, which involve millions of single nucleotide polymorphism (SNPs), where some of the SNPs interact epistatically to affect disease status. Towards determining these interacting SNPs, researchers developed techniques that addressed this specific problem. However, the problem is more general, and so these techniques are applicable to other problems concerning interactions. A difficulty with many of these techniques is that they do not distinguish whether a learned interaction is actually an interaction or whether it involves several variables with strong marginal effects. METHODOLOGY/FINDINGS: We address this problem using information gain and Bayesian network scoring. First, we identify candidate interactions by determining whether together variables provide more information than they do separately. Then we use Bayesian network scoring to see if a candidate interaction really is a likely model. Our strategy is called MBS-IGain. Using 100 simulated datasets and a real GWAS Alzheimer’s dataset, we investigated the performance of MBS-IGain. CONCLUSIONS/SIGNIFICANCE: When analyzing the simulated datasets, MBS-IGain substantially out-performed nine previous methods at locating interacting predictors, and at identifying interactions exactly. When analyzing the real Alzheimer’s dataset, we obtained new results and results that substantiated previous findings. We conclude that MBS-IGain is highly effective at finding interactions in high-dimensional datasets. This result is significant because we have increasingly abundant high-dimensional data in many domains, and to learn causes and perform prediction/classification using these data, we often must first identify interactions. Public Library of Science 2015-12-01 /pmc/articles/PMC4666609/ /pubmed/26624895 http://dx.doi.org/10.1371/journal.pone.0143247 Text en © 2015 Jiang et al http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
spellingShingle Research Article
Jiang, Xia
Jao, Jeremy
Neapolitan, Richard
Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title_full Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title_fullStr Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title_full_unstemmed Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title_short Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring
title_sort learning predictive interactions using information gain and bayesian network scoring
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4666609/
https://www.ncbi.nlm.nih.gov/pubmed/26624895
http://dx.doi.org/10.1371/journal.pone.0143247
work_keys_str_mv AT jiangxia learningpredictiveinteractionsusinginformationgainandbayesiannetworkscoring
AT jaojeremy learningpredictiveinteractionsusinginformationgainandbayesiannetworkscoring
AT neapolitanrichard learningpredictiveinteractionsusinginformationgainandbayesiannetworkscoring