Publication: Min-Uncertainty & Max-Certainty Criteria of Neighborhood Rough-Mutual Feature Selection
Submitted Date
Received Date
Accepted Date
Issued Date
2017
Copyright Date
Announcement No.
Application No.
Patent No.
Valid Date
Resource Type
Edition
Resource Version
Language
en
File Type
No. of Pages/File Size
ISBN
ISSN
2228-835X
eISSN
DOI
Scopus ID
WOS ID
Pubmed ID
arXiv ID
item.page.harrt.identifier.callno
Other identifier(s)
Journal Title
Walailak Journal of Science and Technology (WJST)
Volume
14
Issue
4
Edition
Start Page
275
End Page
297
Access Rights
Access Status
Rights
Copyright (c) 2016 Walailak Journal of Science and Technology (WJST)
Rights Holder(s)
Physical Location
Bibliographic Citation
Research Projects
Organizational Units
Authors
Journal Issue
Title
Min-Uncertainty & Max-Certainty Criteria of Neighborhood Rough-Mutual Feature Selection
Alternative Title(s)
Author’s Affiliation
Author's E-mail
Editor(s)
Editor’s Affiliation
Corresponding person(s)
Creator(s)
Compiler
Advisor(s)
Illustrator(s)
Applicant(s)
Inventor(s)
Issuer
Assignee
Other Contributor(s)
Series
Has Part
Abstract
Feature Selection (FS) is viewed as an important preprocessing step for pattern recognition; machine learning; and data mining. Most existing FS methods based on rough set theory use the dependency function for evaluating the goodness of a feature subset. However; these FS methods may unsuccessfully be applied on dataset with noise; which determine only information from a positive region but neglect a boundary region. This paper proposes a criterion of the maximal lower approximation information (Max-Certainty) and minimal boundary region information (Min-Uncertainty); based on neighborhood rough set and mutual information for evaluating the goodness of a feature subset. We combine this proposed criterion with neighborhood rough set; which is directly applicable to numerical and heterogeneous features; without involving a discretization of numerical features. Comparing it with the rough set based approaches; our proposed method improves accuracy over various experimental data sets. Experimental results illustrate that much valuable information can be extracted by using this idea. This proposed technique is demonstrated on discrete; continuous; and heterogeneous data; and is compared with other FS methods in terms of subset size and classification accuracy.