Tampere University of Technology

TUTCRIS Research Portal

Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders

Research output: Contribution to journalArticleScientificpeer-review

Standard

Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders. / Smolander, Johannes; Dehmer, Matthias; Emmert-Streib, Frank.

In: FEBS Open Bio, Vol. 9, No. 7, 01.07.2019, p. 1232-1248.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Smolander, Johannes ; Dehmer, Matthias ; Emmert-Streib, Frank. / Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders. In: FEBS Open Bio. 2019 ; Vol. 9, No. 7. pp. 1232-1248.

Bibtex - Download

@article{6f7eebd04ceb48f6a244ecaef19745d0,
title = "Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders",
abstract = "Genomics data provide great opportunities for translational research and the clinical practice, for example, for predicting disease stages. However, the classification of such data is a challenging task due to their high dimensionality, noise, and heterogeneity. In recent years, deep learning classifiers generated much interest, but due to their complexity, so far, little is known about the utility of this method for genomics. In this paper, we address this problem by studying a computational diagnostics task by classification of breast cancer and inflammatory bowel disease patients based on high-dimensional gene expression data. We provide a comprehensive analysis of the classification performance of deep belief networks (DBNs) in dependence on its multiple model parameters and in comparison with support vector machines (SVMs). Furthermore, we investigate combined classifiers that integrate DBNs with SVMs. Such a classifier utilizes a DBN as representation learner forming the input for a SVM. Overall, our results provide guidelines for the complex usage of DBN for classifying gene expression data from complex diseases.",
keywords = "artificial intelligence, deep belief network, deep learning, genomics, neural networks, support vector machine",
author = "Johannes Smolander and Matthias Dehmer and Frank Emmert-Streib",
note = "int=comp,{"}Smolander, Johannes{"}",
year = "2019",
month = "7",
day = "1",
doi = "10.1002/2211-5463.12652",
language = "English",
volume = "9",
pages = "1232--1248",
journal = "FEBS Open Bio",
issn = "2211-5463",
publisher = "Wiley-Blackwell",
number = "7",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders

AU - Smolander, Johannes

AU - Dehmer, Matthias

AU - Emmert-Streib, Frank

N1 - int=comp,"Smolander, Johannes"

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Genomics data provide great opportunities for translational research and the clinical practice, for example, for predicting disease stages. However, the classification of such data is a challenging task due to their high dimensionality, noise, and heterogeneity. In recent years, deep learning classifiers generated much interest, but due to their complexity, so far, little is known about the utility of this method for genomics. In this paper, we address this problem by studying a computational diagnostics task by classification of breast cancer and inflammatory bowel disease patients based on high-dimensional gene expression data. We provide a comprehensive analysis of the classification performance of deep belief networks (DBNs) in dependence on its multiple model parameters and in comparison with support vector machines (SVMs). Furthermore, we investigate combined classifiers that integrate DBNs with SVMs. Such a classifier utilizes a DBN as representation learner forming the input for a SVM. Overall, our results provide guidelines for the complex usage of DBN for classifying gene expression data from complex diseases.

AB - Genomics data provide great opportunities for translational research and the clinical practice, for example, for predicting disease stages. However, the classification of such data is a challenging task due to their high dimensionality, noise, and heterogeneity. In recent years, deep learning classifiers generated much interest, but due to their complexity, so far, little is known about the utility of this method for genomics. In this paper, we address this problem by studying a computational diagnostics task by classification of breast cancer and inflammatory bowel disease patients based on high-dimensional gene expression data. We provide a comprehensive analysis of the classification performance of deep belief networks (DBNs) in dependence on its multiple model parameters and in comparison with support vector machines (SVMs). Furthermore, we investigate combined classifiers that integrate DBNs with SVMs. Such a classifier utilizes a DBN as representation learner forming the input for a SVM. Overall, our results provide guidelines for the complex usage of DBN for classifying gene expression data from complex diseases.

KW - artificial intelligence

KW - deep belief network

KW - deep learning

KW - genomics

KW - neural networks

KW - support vector machine

U2 - 10.1002/2211-5463.12652

DO - 10.1002/2211-5463.12652

M3 - Article

VL - 9

SP - 1232

EP - 1248

JO - FEBS Open Bio

JF - FEBS Open Bio

SN - 2211-5463

IS - 7

ER -