Design of Convolutional Fuzzy Neural Network Classifiers

Jiying Men; Wei Huang; Jinsong Wang

doi:10.1109/ICAICE51518.2020.00065

Design of Convolutional Fuzzy Neural Network Classifiers

Jiying Men, Wei Huang^*, Jinsong Wang

^*Corresponding author for this work

Tianjin University of Technology

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

Abstract

In this paper, we propose a convolutional fuzzy neural network classification to alleviate the problems of processing high-dimensional data and low computational efficiency in traditional convolutional neural networks. The model proposes a convolution fuzzy C-means algorithm, in the meanwhile uses the L2-norm regularization method to estimate parameters, so that it has better generalization ability. The experimental results indicate that the proposed CFNNCs have excellent performance in classification accuracy than classic classification models such as SVM, RVM, KNN, and the experimental accuracy can be maintained above 90%.

Original language	English
Title of host publication	Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	306-309
Number of pages	4
ISBN (Electronic)	9781728191461
DOIs	https://doi.org/10.1109/ICAICE51518.2020.00065
Publication status	Published - Oct 2020
Externally published	Yes
Event	2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020 - Beijing, China Duration: 23 Oct 2020 → 25 Oct 2020

Publication series

Name	Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020

Conference

Conference	2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020
Country/Territory	China
City	Beijing
Period	23/10/20 → 25/10/20

Keywords

L2-norm regularization
convolutional neural network
fuzzy rules

Access to Document

10.1109/ICAICE51518.2020.00065

Cite this

Men, J., Huang, W., & Wang, J. (2020). Design of Convolutional Fuzzy Neural Network Classifiers. In Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020 (pp. 306-309). Article 9361333 (Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICAICE51518.2020.00065

Men, Jiying ; Huang, Wei ; Wang, Jinsong. / Design of Convolutional Fuzzy Neural Network Classifiers. Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 306-309 (Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020).

@inproceedings{c4b84c0549f0407bbf43462d58ef025a,

title = "Design of Convolutional Fuzzy Neural Network Classifiers",

abstract = "In this paper, we propose a convolutional fuzzy neural network classification to alleviate the problems of processing high-dimensional data and low computational efficiency in traditional convolutional neural networks. The model proposes a convolution fuzzy C-means algorithm, in the meanwhile uses the L2-norm regularization method to estimate parameters, so that it has better generalization ability. The experimental results indicate that the proposed CFNNCs have excellent performance in classification accuracy than classic classification models such as SVM, RVM, KNN, and the experimental accuracy can be maintained above 90%.",

keywords = "L2-norm regularization, convolutional neural network, fuzzy rules",

author = "Jiying Men and Wei Huang and Jinsong Wang",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020 ; Conference date: 23-10-2020 Through 25-10-2020",

year = "2020",

month = oct,

doi = "10.1109/ICAICE51518.2020.00065",

language = "English",

series = "Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020",

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

pages = "306--309",

booktitle = "Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020",

address = "United States",

}

Men, J, Huang, W & Wang, J 2020, Design of Convolutional Fuzzy Neural Network Classifiers. in Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020., 9361333, Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020, Institute of Electrical and Electronics Engineers Inc., pp. 306-309, 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020, Beijing, China, 23/10/20. https://doi.org/10.1109/ICAICE51518.2020.00065

Design of Convolutional Fuzzy Neural Network Classifiers. / Men, Jiying; Huang, Wei; Wang, Jinsong.
Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020. Institute of Electrical and Electronics Engineers Inc., 2020. p. 306-309 9361333 (Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020).

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

TY - GEN

T1 - Design of Convolutional Fuzzy Neural Network Classifiers

AU - Men, Jiying

AU - Huang, Wei

AU - Wang, Jinsong

PY - 2020/10

Y1 - 2020/10

N2 - In this paper, we propose a convolutional fuzzy neural network classification to alleviate the problems of processing high-dimensional data and low computational efficiency in traditional convolutional neural networks. The model proposes a convolution fuzzy C-means algorithm, in the meanwhile uses the L2-norm regularization method to estimate parameters, so that it has better generalization ability. The experimental results indicate that the proposed CFNNCs have excellent performance in classification accuracy than classic classification models such as SVM, RVM, KNN, and the experimental accuracy can be maintained above 90%.

AB - In this paper, we propose a convolutional fuzzy neural network classification to alleviate the problems of processing high-dimensional data and low computational efficiency in traditional convolutional neural networks. The model proposes a convolution fuzzy C-means algorithm, in the meanwhile uses the L2-norm regularization method to estimate parameters, so that it has better generalization ability. The experimental results indicate that the proposed CFNNCs have excellent performance in classification accuracy than classic classification models such as SVM, RVM, KNN, and the experimental accuracy can be maintained above 90%.

KW - L2-norm regularization

KW - convolutional neural network

KW - fuzzy rules

UR - http://www.scopus.com/inward/record.url?scp=85102660704&partnerID=8YFLogxK

U2 - 10.1109/ICAICE51518.2020.00065

DO - 10.1109/ICAICE51518.2020.00065

M3 - Conference contribution

AN - SCOPUS:85102660704

T3 - Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020

SP - 306

EP - 309

BT - Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020

Y2 - 23 October 2020 through 25 October 2020

ER -

Men J, Huang W, Wang J. Design of Convolutional Fuzzy Neural Network Classifiers. In Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 306-309. 9361333. (Proceedings - 2020 International Conference on Artificial Intelligence and Computer Engineering, ICAICE 2020). doi: 10.1109/ICAICE51518.2020.00065

Design of Convolutional Fuzzy Neural Network Classifiers

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this