Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection

Cui Na Jiao; Tian Ru Wu; Jin Xing Liu; Xiang Zhen Kong

doi:10.1007/978-981-15-8760-3_8

Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection

Cui Na Jiao
, Tian Ru Wu
, Jin Xing Liu
, Xiang Zhen Kong

Qufu Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Non-negative Matrix Factorization (NMF) algorithm is a useful method for data dimensionality reduction, which is performed with the Euclidean distance. However, the basic NMF only assumes that data will be destroyed by Gaussian noise. It ignores both the intrinsic geometrical structure and the influence of sparse noises existing in the gene expression data. To enhance the robustness of the NMF, a novel method called Hyper-graph Robust Non-negative Matrix Factorization (HRNMF) is proposed for cancer sample clustering and feature selection. The merits of the HRNMF mainly consist of two aspects. Firstly, the L_{2, 1}-norm is combined with the objective function, which can effectively handle noise and outliers. Secondly, the manifold information and sparsity are also considered so we add the hyper-graph regularization term and sparse constraints to the error function. It can effectively preserve the geometric structure and enhance matrix sparsity. Experiments on Cancer Genome Atlas (TCGA) gene expression data have demonstrated that HRNMF performs better than other advanced methods in cancer sample clustering and feature selection.

Original language	English
Title of host publication	Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers
Editors	Henry Han, Tie Wei, Wenbin Liu, Fei Han
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	112-125
Number of pages	14
ISBN (Print)	9789811587597
DOIs	https://doi.org/10.1007/978-981-15-8760-3_8
State	Published - 2020
Externally published	Yes
Event	3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019 - Nanning, China Duration: 22 Jun 2019 → 24 Jun 2019

Publication series

Name	Communications in Computer and Information Science
Volume	1099 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019
Country/Territory	China
City	Nanning
Period	22/06/19 → 24/06/19

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Hyper-graph regularization
L-norm
Non-negative Matrix Factorization
Robustness
Sample clustering

Access to Document

10.1007/978-981-15-8760-3_8

Cite this

Jiao, C. N., Wu, T. R., Liu, J. X., & Kong, X. Z. (2020). Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection. In H. Han, T. Wei, W. Liu, & F. Han (Eds.), Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers (pp. 112-125). (Communications in Computer and Information Science; Vol. 1099 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-15-8760-3_8

Jiao, Cui Na ; Wu, Tian Ru ; Liu, Jin Xing et al. / Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection. Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers. editor / Henry Han ; Tie Wei ; Wenbin Liu ; Fei Han. Springer Science and Business Media Deutschland GmbH, 2020. pp. 112-125 (Communications in Computer and Information Science).

@inproceedings{d25ec96251084c2a886980541a19fe8a,

title = "Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection",

abstract = "Non-negative Matrix Factorization (NMF) algorithm is a useful method for data dimensionality reduction, which is performed with the Euclidean distance. However, the basic NMF only assumes that data will be destroyed by Gaussian noise. It ignores both the intrinsic geometrical structure and the influence of sparse noises existing in the gene expression data. To enhance the robustness of the NMF, a novel method called Hyper-graph Robust Non-negative Matrix Factorization (HRNMF) is proposed for cancer sample clustering and feature selection. The merits of the HRNMF mainly consist of two aspects. Firstly, the L2, 1-norm is combined with the objective function, which can effectively handle noise and outliers. Secondly, the manifold information and sparsity are also considered so we add the hyper-graph regularization term and sparse constraints to the error function. It can effectively preserve the geometric structure and enhance matrix sparsity. Experiments on Cancer Genome Atlas (TCGA) gene expression data have demonstrated that HRNMF performs better than other advanced methods in cancer sample clustering and feature selection.",

keywords = "Hyper-graph regularization, L-norm, Non-negative Matrix Factorization, Robustness, Sample clustering",

author = "Jiao, \{Cui Na\} and Wu, \{Tian Ru\} and Liu, \{Jin Xing\} and Kong, \{Xiang Zhen\}",

note = "Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd 2020.; 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019 ; Conference date: 22-06-2019 Through 24-06-2019",

year = "2020",

doi = "10.1007/978-981-15-8760-3\_8",

language = "英语",

isbn = "9789811587597",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "112--125",

editor = "Henry Han and Tie Wei and Wenbin Liu and Fei Han",

booktitle = "Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers",

}

Jiao, CN, Wu, TR, Liu, JX & Kong, XZ 2020, Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection. in H Han, T Wei, W Liu & F Han (eds), Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers. Communications in Computer and Information Science, vol. 1099 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 112-125, 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Nanning, China, 22/06/19. https://doi.org/10.1007/978-981-15-8760-3_8

Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection. / Jiao, Cui Na; Wu, Tian Ru; Liu, Jin Xing et al.
Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers. ed. / Henry Han; Tie Wei; Wenbin Liu; Fei Han. Springer Science and Business Media Deutschland GmbH, 2020. p. 112-125 (Communications in Computer and Information Science; Vol. 1099 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection

AU - Jiao, Cui Na

AU - Wu, Tian Ru

AU - Liu, Jin Xing

AU - Kong, Xiang Zhen

N1 - Publisher Copyright: © Springer Nature Singapore Pte Ltd 2020.

PY - 2020

Y1 - 2020

N2 - Non-negative Matrix Factorization (NMF) algorithm is a useful method for data dimensionality reduction, which is performed with the Euclidean distance. However, the basic NMF only assumes that data will be destroyed by Gaussian noise. It ignores both the intrinsic geometrical structure and the influence of sparse noises existing in the gene expression data. To enhance the robustness of the NMF, a novel method called Hyper-graph Robust Non-negative Matrix Factorization (HRNMF) is proposed for cancer sample clustering and feature selection. The merits of the HRNMF mainly consist of two aspects. Firstly, the L2, 1-norm is combined with the objective function, which can effectively handle noise and outliers. Secondly, the manifold information and sparsity are also considered so we add the hyper-graph regularization term and sparse constraints to the error function. It can effectively preserve the geometric structure and enhance matrix sparsity. Experiments on Cancer Genome Atlas (TCGA) gene expression data have demonstrated that HRNMF performs better than other advanced methods in cancer sample clustering and feature selection.

AB - Non-negative Matrix Factorization (NMF) algorithm is a useful method for data dimensionality reduction, which is performed with the Euclidean distance. However, the basic NMF only assumes that data will be destroyed by Gaussian noise. It ignores both the intrinsic geometrical structure and the influence of sparse noises existing in the gene expression data. To enhance the robustness of the NMF, a novel method called Hyper-graph Robust Non-negative Matrix Factorization (HRNMF) is proposed for cancer sample clustering and feature selection. The merits of the HRNMF mainly consist of two aspects. Firstly, the L2, 1-norm is combined with the objective function, which can effectively handle noise and outliers. Secondly, the manifold information and sparsity are also considered so we add the hyper-graph regularization term and sparse constraints to the error function. It can effectively preserve the geometric structure and enhance matrix sparsity. Experiments on Cancer Genome Atlas (TCGA) gene expression data have demonstrated that HRNMF performs better than other advanced methods in cancer sample clustering and feature selection.

KW - Hyper-graph regularization

KW - L-norm

KW - Non-negative Matrix Factorization

KW - Robustness

KW - Sample clustering

UR - https://www.scopus.com/pages/publications/85092173230

U2 - 10.1007/978-981-15-8760-3_8

DO - 10.1007/978-981-15-8760-3_8

M3 - 会议稿件

AN - SCOPUS:85092173230

SN - 9789811587597

T3 - Communications in Computer and Information Science

SP - 112

EP - 125

BT - Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers

A2 - Han, Henry

A2 - Wei, Tie

A2 - Liu, Wenbin

A2 - Han, Fei

PB - Springer Science and Business Media Deutschland GmbH

T2 - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019

Y2 - 22 June 2019 through 24 June 2019

ER -

Jiao CN, Wu TR, Liu JX, Kong XZ. Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection. In Han H, Wei T, Liu W, Han F, editors, Recent Advances in Data Science - 3rd International Conference on Data Science, Medicine, and Bioinformatics, IDMB 2019, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2020. p. 112-125. (Communications in Computer and Information Science). doi: 10.1007/978-981-15-8760-3_8

Hyper-graph robust non-negative matrix factorization method for cancer sample clustering and feature selection

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this