Feature selection with data field

Hanning Yuan; Shuliang Wang; Ying Li; Jinghua Fan

Feature selection with data field

Hanning Yuan
, Shuliang Wang^*
, Ying Li
, Jinghua Fan

^*Corresponding author for this work

Beijing Institute of Technology
Wuhan University

Research output: Contribution to journal › Article › peer-review

8 Citations (Scopus)

Abstract

A new feature selection method is proposed for high-dimensional data clustering on the basis of data field. With the potential entropy to evaluate the importance of feature subsets, features are filtered by removing unimportant features or noises from the original datasets. Experiments show that the proposed method can sharply reduce the number of dimensions and effectively improve the clustering performance on WDBC dataset.

Original language	English
Pages (from-to)	661-665
Number of pages	5
Journal	Chinese Journal of Electronics
Volume	23
Issue number	4
Publication status	Published - 1 Oct 2014
Externally published	Yes

Keywords

Data field
Dimension reduction
Feature selection
High-dimensional objects
Potential entropy

Cite this

@article{86827e06c2fd4cf7ac9f8ccfc0b0b917,

title = "Feature selection with data field",

abstract = "A new feature selection method is proposed for high-dimensional data clustering on the basis of data field. With the potential entropy to evaluate the importance of feature subsets, features are filtered by removing unimportant features or noises from the original datasets. Experiments show that the proposed method can sharply reduce the number of dimensions and effectively improve the clustering performance on WDBC dataset.",

keywords = "Data field, Dimension reduction, Feature selection, High-dimensional objects, Potential entropy",

author = "Hanning Yuan and Shuliang Wang and Ying Li and Jinghua Fan",

year = "2014",

month = oct,

day = "1",

language = "English",

volume = "23",

pages = "661--665",

journal = "Chinese Journal of Electronics",

issn = "1022-4653",

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

number = "4",

}

TY - JOUR

T1 - Feature selection with data field

AU - Yuan, Hanning

AU - Wang, Shuliang

AU - Li, Ying

AU - Fan, Jinghua

PY - 2014/10/1

Y1 - 2014/10/1

N2 - A new feature selection method is proposed for high-dimensional data clustering on the basis of data field. With the potential entropy to evaluate the importance of feature subsets, features are filtered by removing unimportant features or noises from the original datasets. Experiments show that the proposed method can sharply reduce the number of dimensions and effectively improve the clustering performance on WDBC dataset.

AB - A new feature selection method is proposed for high-dimensional data clustering on the basis of data field. With the potential entropy to evaluate the importance of feature subsets, features are filtered by removing unimportant features or noises from the original datasets. Experiments show that the proposed method can sharply reduce the number of dimensions and effectively improve the clustering performance on WDBC dataset.

KW - Data field

KW - Dimension reduction

KW - Feature selection

KW - High-dimensional objects

KW - Potential entropy

UR - https://www.scopus.com/pages/publications/84907506434

M3 - Article

AN - SCOPUS:84907506434

SN - 1022-4653

VL - 23

SP - 661

EP - 665

JO - Chinese Journal of Electronics

JF - Chinese Journal of Electronics

IS - 4

ER -

Feature selection with data field

Abstract

Keywords

Other files and links

Fingerprint

Cite this