Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations

Hao Huang; Shinjae Yoo; Dantong Yu; Hong Qin

doi:10.1109/ICDM.2014.88

Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations

Hao Huang
, Shinjae Yoo
, Dantong Yu
, Hong Qin

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

2 引用（Scopus）

摘要

Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features. We propose a noise-resistant feature selection algorithm by capturing multi-perspective correlations. Our proposed approach, called Noise-Resistant Unsupervised Feature Selection (NRFS), is based on multi-perspective correlation that reflects the importance of feature with respect to noise-resistant representative instances and various global trends from spectral decomposition. In this way, the model concisely captures a wide variety of local patterns. Experimental results demonstrate the effectiveness of our algorithm.

源语言	英语
主期刊名	Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014
编辑	Ravi Kumar, Hannu Toivonen, Jian Pei, Joshua Zhexue Huang, Xindong Wu
出版商	Institute of Electrical and Electronics Engineers Inc.
页	210-219
页数	10
版本	January
ISBN（电子版）	9781479943029
DOI	https://doi.org/10.1109/ICDM.2014.88
出版状态	已出版 - 1 1月 2014
已对外发布	是
活动	14th IEEE International Conference on Data Mining, ICDM 2014 - Shenzhen, 中国期限: 14 12月 2014 → 17 12月 2014

出版系列

姓名	Proceedings - IEEE International Conference on Data Mining, ICDM
编号	January
卷	2015-January
ISSN（印刷版）	1550-4786

会议

会议	14th IEEE International Conference on Data Mining, ICDM 2014
国家/地区	中国
市	Shenzhen
时期	14/12/14 → 17/12/14

访问文件

10.1109/ICDM.2014.88

其它文件与链接

链接到 Scopus 的出版物

引用此

Huang, H., Yoo, S., Yu, D., & Qin, H. (2014). Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations. 在 R. Kumar, H. Toivonen, J. Pei, J. Zhexue Huang, & X. Wu (编辑), Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014 (January 编辑, 页码 210-219). 文章 7023338 (Proceedings - IEEE International Conference on Data Mining, ICDM; 卷 2015-January, 号码 January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDM.2014.88

Huang, Hao ; Yoo, Shinjae ; Yu, Dantong 等. / Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations. Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014. 编辑 / Ravi Kumar ; Hannu Toivonen ; Jian Pei ; Joshua Zhexue Huang ; Xindong Wu. January. 编辑 Institute of Electrical and Electronics Engineers Inc., 2014. 页码 210-219 (Proceedings - IEEE International Conference on Data Mining, ICDM; January).

@inproceedings{1a1c2af0a6f743379f4f1b8da63b12b2,

title = "Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations",

abstract = "Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features. We propose a noise-resistant feature selection algorithm by capturing multi-perspective correlations. Our proposed approach, called Noise-Resistant Unsupervised Feature Selection (NRFS), is based on multi-perspective correlation that reflects the importance of feature with respect to noise-resistant representative instances and various global trends from spectral decomposition. In this way, the model concisely captures a wide variety of local patterns. Experimental results demonstrate the effectiveness of our algorithm.",

author = "Hao Huang and Shinjae Yoo and Dantong Yu and Hong Qin",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 14th IEEE International Conference on Data Mining, ICDM 2014 ; Conference date: 14-12-2014 Through 17-12-2014",

year = "2014",

month = jan,

day = "1",

doi = "10.1109/ICDM.2014.88",

language = "英语",

series = "Proceedings - IEEE International Conference on Data Mining, ICDM",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "January",

pages = "210--219",

editor = "Ravi Kumar and Hannu Toivonen and Jian Pei and \{Zhexue Huang\}, Joshua and Xindong Wu",

booktitle = "Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014",

address = "美国",

edition = "January",

}

Huang, H, Yoo, S, Yu, D & Qin, H 2014, Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations. 在 R Kumar, H Toivonen, J Pei, J Zhexue Huang & X Wu (编辑), Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014. January 编辑, 7023338, Proceedings - IEEE International Conference on Data Mining, ICDM, 号码 January, 卷 2015-January, Institute of Electrical and Electronics Engineers Inc., 页码 210-219, 14th IEEE International Conference on Data Mining, ICDM 2014, Shenzhen, 中国, 14/12/14. https://doi.org/10.1109/ICDM.2014.88

Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations. / Huang, Hao; Yoo, Shinjae; Yu, Dantong 等.
Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014. 编辑 / Ravi Kumar; Hannu Toivonen; Jian Pei; Joshua Zhexue Huang; Xindong Wu. January. 编辑 Institute of Electrical and Electronics Engineers Inc., 2014. 页码 210-219 7023338 (Proceedings - IEEE International Conference on Data Mining, ICDM; 卷 2015-January, 号码 January).

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

TY - GEN

T1 - Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations

AU - Huang, Hao

AU - Yoo, Shinjae

AU - Yu, Dantong

AU - Qin, Hong

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features. We propose a noise-resistant feature selection algorithm by capturing multi-perspective correlations. Our proposed approach, called Noise-Resistant Unsupervised Feature Selection (NRFS), is based on multi-perspective correlation that reflects the importance of feature with respect to noise-resistant representative instances and various global trends from spectral decomposition. In this way, the model concisely captures a wide variety of local patterns. Experimental results demonstrate the effectiveness of our algorithm.

AB - Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features. We propose a noise-resistant feature selection algorithm by capturing multi-perspective correlations. Our proposed approach, called Noise-Resistant Unsupervised Feature Selection (NRFS), is based on multi-perspective correlation that reflects the importance of feature with respect to noise-resistant representative instances and various global trends from spectral decomposition. In this way, the model concisely captures a wide variety of local patterns. Experimental results demonstrate the effectiveness of our algorithm.

UR - https://www.scopus.com/pages/publications/84936932465

U2 - 10.1109/ICDM.2014.88

DO - 10.1109/ICDM.2014.88

M3 - 会议稿件

AN - SCOPUS:84936932465

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 210

EP - 219

BT - Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014

A2 - Kumar, Ravi

A2 - Toivonen, Hannu

A2 - Pei, Jian

A2 - Zhexue Huang, Joshua

A2 - Wu, Xindong

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 14th IEEE International Conference on Data Mining, ICDM 2014

Y2 - 14 December 2014 through 17 December 2014

ER -

Huang H, Yoo S, Yu D, Qin H. Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations. 在 Kumar R, Toivonen H, Pei J, Zhexue Huang J, Wu X, 编辑, Proceedings - 14th IEEE International Conference on Data Mining, ICDM 2014. January 编辑 Institute of Electrical and Electronics Engineers Inc. 2014. 页码 210-219. 7023338. (Proceedings - IEEE International Conference on Data Mining, ICDM; January). doi: 10.1109/ICDM.2014.88

Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations

摘要

出版系列

会议

访问文件

其它文件与链接

指纹

引用此