Fuzzy kernel alignment with application to attribute reduction of heterogeneous data

Linlin Chen, Degang Chen, H. Wang

Research output: Contribution to journalArticle

Abstract

Fuzzy similarity relation is a function to measure the similarity between two samples. It is widely used to learn knowledge under the framework of fuzzy machine learning, and the selection of a suitable fuzzy similarity relation is obviously important for the learning task. It has been pointed out that fuzzy similarity relations can be brought into the framework of kernel functions in machine learning. This fact motivates us to study fuzzy similarity relation selection for fuzzy machine learning utilizing kernel selection methods in machine learning. Kernel alignment is a kernel selection method that is effective and has low computational complexity. In this paper we present novel methods for fuzzy similarity relation selection based on kernel alignment, and their use in attribution reduction for heterogeneous data. Firstly, we define an ideal kernel for classification problems, based on which a novel fuzzy kernel alignment model is proposed. Secondly, we present a method for fuzzy similarity relation selection based on the minimization of fuzzy alignment between the defined ideal kernel and a kernel for the learning problem at hand. In order to show the correctness of this selection method, we prove that the lower bound of the classification accuracy of a support vector machine will increase with the decrease of the fuzzy alignment value. Furthermore, we apply the proposed fuzzy similarity relation selection to attribute reduction for heterogeneous data. Finally, we present experimental results to show that the proposed method of fuzzy similarity relation selection based on fuzzy kernel alignment is effective.
LanguageEnglish
Pages1469-1478
JournalIEEE Transactions on Fuzzy Systems
Volume27
Issue number7
Early online date12 Nov 2018
DOIs
Publication statusE-pub ahead of print - 12 Nov 2018

Fingerprint

Attribute Reduction
Similarity Relation
Fuzzy Relation
Alignment
kernel
Learning systems
Machine Learning
Support vector machines
Computational complexity
Kernel Function
Classification Problems
Low Complexity
Support Vector Machine
Correctness
Computational Complexity
Lower bound
Decrease

Cite this

@article{8c27f9617f664cd3816f51daf90b17fa,
title = "Fuzzy kernel alignment with application to attribute reduction of heterogeneous data",
abstract = "Fuzzy similarity relation is a function to measure the similarity between two samples. It is widely used to learn knowledge under the framework of fuzzy machine learning, and the selection of a suitable fuzzy similarity relation is obviously important for the learning task. It has been pointed out that fuzzy similarity relations can be brought into the framework of kernel functions in machine learning. This fact motivates us to study fuzzy similarity relation selection for fuzzy machine learning utilizing kernel selection methods in machine learning. Kernel alignment is a kernel selection method that is effective and has low computational complexity. In this paper we present novel methods for fuzzy similarity relation selection based on kernel alignment, and their use in attribution reduction for heterogeneous data. Firstly, we define an ideal kernel for classification problems, based on which a novel fuzzy kernel alignment model is proposed. Secondly, we present a method for fuzzy similarity relation selection based on the minimization of fuzzy alignment between the defined ideal kernel and a kernel for the learning problem at hand. In order to show the correctness of this selection method, we prove that the lower bound of the classification accuracy of a support vector machine will increase with the decrease of the fuzzy alignment value. Furthermore, we apply the proposed fuzzy similarity relation selection to attribute reduction for heterogeneous data. Finally, we present experimental results to show that the proposed method of fuzzy similarity relation selection based on fuzzy kernel alignment is effective.",
author = "Linlin Chen and Degang Chen and H. Wang",
year = "2018",
month = "11",
day = "12",
doi = "10.1109/TFUZZ.2018.2880933",
language = "English",
volume = "27",
pages = "1469--1478",
journal = "IEEE Transactions on Fuzzy Systems",
issn = "1063-6706",
number = "7",

}

Fuzzy kernel alignment with application to attribute reduction of heterogeneous data. / Chen, Linlin; Chen, Degang; Wang, H.

In: IEEE Transactions on Fuzzy Systems, Vol. 27, No. 7, 12.11.2018, p. 1469-1478.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fuzzy kernel alignment with application to attribute reduction of heterogeneous data

AU - Chen, Linlin

AU - Chen, Degang

AU - Wang, H.

PY - 2018/11/12

Y1 - 2018/11/12

N2 - Fuzzy similarity relation is a function to measure the similarity between two samples. It is widely used to learn knowledge under the framework of fuzzy machine learning, and the selection of a suitable fuzzy similarity relation is obviously important for the learning task. It has been pointed out that fuzzy similarity relations can be brought into the framework of kernel functions in machine learning. This fact motivates us to study fuzzy similarity relation selection for fuzzy machine learning utilizing kernel selection methods in machine learning. Kernel alignment is a kernel selection method that is effective and has low computational complexity. In this paper we present novel methods for fuzzy similarity relation selection based on kernel alignment, and their use in attribution reduction for heterogeneous data. Firstly, we define an ideal kernel for classification problems, based on which a novel fuzzy kernel alignment model is proposed. Secondly, we present a method for fuzzy similarity relation selection based on the minimization of fuzzy alignment between the defined ideal kernel and a kernel for the learning problem at hand. In order to show the correctness of this selection method, we prove that the lower bound of the classification accuracy of a support vector machine will increase with the decrease of the fuzzy alignment value. Furthermore, we apply the proposed fuzzy similarity relation selection to attribute reduction for heterogeneous data. Finally, we present experimental results to show that the proposed method of fuzzy similarity relation selection based on fuzzy kernel alignment is effective.

AB - Fuzzy similarity relation is a function to measure the similarity between two samples. It is widely used to learn knowledge under the framework of fuzzy machine learning, and the selection of a suitable fuzzy similarity relation is obviously important for the learning task. It has been pointed out that fuzzy similarity relations can be brought into the framework of kernel functions in machine learning. This fact motivates us to study fuzzy similarity relation selection for fuzzy machine learning utilizing kernel selection methods in machine learning. Kernel alignment is a kernel selection method that is effective and has low computational complexity. In this paper we present novel methods for fuzzy similarity relation selection based on kernel alignment, and their use in attribution reduction for heterogeneous data. Firstly, we define an ideal kernel for classification problems, based on which a novel fuzzy kernel alignment model is proposed. Secondly, we present a method for fuzzy similarity relation selection based on the minimization of fuzzy alignment between the defined ideal kernel and a kernel for the learning problem at hand. In order to show the correctness of this selection method, we prove that the lower bound of the classification accuracy of a support vector machine will increase with the decrease of the fuzzy alignment value. Furthermore, we apply the proposed fuzzy similarity relation selection to attribute reduction for heterogeneous data. Finally, we present experimental results to show that the proposed method of fuzzy similarity relation selection based on fuzzy kernel alignment is effective.

U2 - 10.1109/TFUZZ.2018.2880933

DO - 10.1109/TFUZZ.2018.2880933

M3 - Article

VL - 27

SP - 1469

EP - 1478

JO - IEEE Transactions on Fuzzy Systems

T2 - IEEE Transactions on Fuzzy Systems

JF - IEEE Transactions on Fuzzy Systems

SN - 1063-6706

IS - 7

ER -