Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis

Yuqing GAO; Wing W. Y. NG; Ting WANG; Sam KWONG

doi:10.1109/ICCICC46617.2019.9146039

Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis

Yuqing GAO, Wing W. Y. NG, Ting WANG, Sam KWONG

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

Abstract

Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively small sample size. In this paper, we utilized Dual Denoising Autoencoder Features (DDAF), which integrates two Denoising Auto-Encoders (DAE) with different activation function to map the features for both minority and majority classes into a better classification representation. The experimental results on four typical microarray datasets show that the DDAF outperforms the Dual Autoencoder Features (DAF) and the Cost-sensitive Oversampling Stacked Denoising Auto-Encoder (CO-SDAE), rendering the robust ability for dimensionality reduction and imbalanced classification.

Original language	English
Title of host publication	Proceedings of the 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019
Publisher	IEEE
Pages	366-371
Number of pages	6
ISBN (Electronic)	9781728114194
ISBN (Print)	9781728104966
DOIs	https://doi.org/10.1109/ICCICC46617.2019.9146039
Publication status	Published - 1 Jul 2019
Externally published	Yes
Event	18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019 - Politecnico di Milano University, Milan, Italy Duration: 23 Jul 2019 → 25 Jul 2019 https://www.iccicc19.polimi.it/

Conference

Conference	18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019
Country/Territory	Italy
City	Milan
Period	23/07/19 → 25/07/19
Internet address	https://www.iccicc19.polimi.it/

Keywords

denoising autoencoder
feature learning
imbalanced classification

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICCICC46617.2019.9146039

Cite this

@inproceedings{175afc375b274c299f214ca41e2e982e,

title = "Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis",

abstract = "Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively small sample size. In this paper, we utilized Dual Denoising Autoencoder Features (DDAF), which integrates two Denoising Auto-Encoders (DAE) with different activation function to map the features for both minority and majority classes into a better classification representation. The experimental results on four typical microarray datasets show that the DDAF outperforms the Dual Autoencoder Features (DAF) and the Cost-sensitive Oversampling Stacked Denoising Auto-Encoder (CO-SDAE), rendering the robust ability for dimensionality reduction and imbalanced classification.",

keywords = "denoising autoencoder, feature learning, imbalanced classification",

author = "Yuqing GAO and NG, {Wing W. Y.} and Ting WANG and Sam KWONG",

year = "2019",

month = jul,

day = "1",

doi = "10.1109/ICCICC46617.2019.9146039",

language = "English",

isbn = "9781728104966",

pages = "366--371",

booktitle = "Proceedings of the 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019",

publisher = "IEEE",

note = "18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019<br/> ; Conference date: 23-07-2019 Through 25-07-2019",

url = "https://www.iccicc19.polimi.it/",

}

GAO, Y, NG, WWY, WANG, T & KWONG, S 2019, Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis. in Proceedings of the 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019. IEEE, pp. 366-371, 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019
, Milan, Italy, 23/07/19. https://doi.org/10.1109/ICCICC46617.2019.9146039

Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis. / GAO, Yuqing; NG, Wing W. Y.; WANG, Ting et al.
Proceedings of the 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019. IEEE, 2019. p. 366-371.

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis

AU - GAO, Yuqing

AU - NG, Wing W. Y.

AU - WANG, Ting

AU - KWONG, Sam

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively small sample size. In this paper, we utilized Dual Denoising Autoencoder Features (DDAF), which integrates two Denoising Auto-Encoders (DAE) with different activation function to map the features for both minority and majority classes into a better classification representation. The experimental results on four typical microarray datasets show that the DDAF outperforms the Dual Autoencoder Features (DAF) and the Cost-sensitive Oversampling Stacked Denoising Auto-Encoder (CO-SDAE), rendering the robust ability for dimensionality reduction and imbalanced classification.

AB - Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively small sample size. In this paper, we utilized Dual Denoising Autoencoder Features (DDAF), which integrates two Denoising Auto-Encoders (DAE) with different activation function to map the features for both minority and majority classes into a better classification representation. The experimental results on four typical microarray datasets show that the DDAF outperforms the Dual Autoencoder Features (DAF) and the Cost-sensitive Oversampling Stacked Denoising Auto-Encoder (CO-SDAE), rendering the robust ability for dimensionality reduction and imbalanced classification.

KW - denoising autoencoder

KW - feature learning

KW - imbalanced classification

UR - http://www.scopus.com/inward/record.url?scp=85094682782&partnerID=8YFLogxK

U2 - 10.1109/ICCICC46617.2019.9146039

DO - 10.1109/ICCICC46617.2019.9146039

M3 - Conference paper (refereed)

SN - 9781728104966

SP - 366

EP - 371

BT - Proceedings of the 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019

PB - IEEE

T2 - 18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019<br/>

Y2 - 23 July 2019 through 25 July 2019

ER -

Dual Denoising Autoencoder Feature Learning for Cancer Diagnosis

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this