Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation

Margarita Terziyska, Yancho Todorov, Maria Dobreva

Research output: Chapter in Book/Report/Conference proceedingChapter or book articleScientificpeer-review

1 Citation (Scopus)

Abstract

In this paper the effectiveness of different error metrics for assessment of the capabilities of an advanced fuzzy-neural architecture are studied. The proposed structure combines the potentials of the Intuitionistic Fuzzy Logic with the simplicity of the Neo-Fuzzy Neuron theory for implementation of robust modeling mechanisms, able to capture uncertain variations in the data space. A major concern when evaluating the performance of such kind of models is the selection of appropriate error metrics in order to assess their potential to capture a wide range of system behaviours. Therefore, different error metrics to evaluate the functional properties of a proposed Intuitionistic Neo-fuzzy network are studied and a comparative analysis in modeling of chaotic time series is made.

Original languageEnglish
Title of host publicationAdvanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers
EditorsMichail Todorov, Ivan Georgiev, Krassimir Georgiev, Ivan Georgiev
PublisherSpringer
Pages185-201
Number of pages17
ISBN (Print)978-331965529-1
DOIs
Publication statusPublished - 1 Jan 2018
MoE publication typeA3 Part of a book or another research book
Event11th Annual Meeting of the Bulgarian Section of the Society for Industrial and Applied Mathematics, BGSIAM 2016 - Sofia, Bulgaria
Duration: 20 Dec 201622 Dec 2016

Publication series

SeriesStudies in Computational Intelligence
Volume728
ISSN1860-949X

Conference

Conference11th Annual Meeting of the Bulgarian Section of the Society for Industrial and Applied Mathematics, BGSIAM 2016
CountryBulgaria
CitySofia
Period20/12/1622/12/16

Fingerprint

Fuzzy neural networks
Network performance
Fuzzy logic
Neurons
Time series

Cite this

Terziyska, M., Todorov, Y., & Dobreva, M. (2018). Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation. In M. Todorov, I. Georgiev, K. Georgiev, & I. Georgiev (Eds.), Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers (pp. 185-201). Springer. Studies in Computational Intelligence, Vol.. 728 https://doi.org/10.1007/978-3-319-65530-7_17
Terziyska, Margarita ; Todorov, Yancho ; Dobreva, Maria. / Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation. Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers. editor / Michail Todorov ; Ivan Georgiev ; Krassimir Georgiev ; Ivan Georgiev. Springer, 2018. pp. 185-201 (Studies in Computational Intelligence, Vol. 728).
@inbook{88966855477a4b039ccc5d5a5871112b,
title = "Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation",
abstract = "In this paper the effectiveness of different error metrics for assessment of the capabilities of an advanced fuzzy-neural architecture are studied. The proposed structure combines the potentials of the Intuitionistic Fuzzy Logic with the simplicity of the Neo-Fuzzy Neuron theory for implementation of robust modeling mechanisms, able to capture uncertain variations in the data space. A major concern when evaluating the performance of such kind of models is the selection of appropriate error metrics in order to assess their potential to capture a wide range of system behaviours. Therefore, different error metrics to evaluate the functional properties of a proposed Intuitionistic Neo-fuzzy network are studied and a comparative analysis in modeling of chaotic time series is made.",
author = "Margarita Terziyska and Yancho Todorov and Maria Dobreva",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-319-65530-7_17",
language = "English",
isbn = "978-331965529-1",
series = "Studies in Computational Intelligence",
publisher = "Springer",
pages = "185--201",
editor = "Michail Todorov and Ivan Georgiev and Krassimir Georgiev and Ivan Georgiev",
booktitle = "Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers",
address = "Germany",

}

Terziyska, M, Todorov, Y & Dobreva, M 2018, Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation. in M Todorov, I Georgiev, K Georgiev & I Georgiev (eds), Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers. Springer, Studies in Computational Intelligence, vol. 728, pp. 185-201, 11th Annual Meeting of the Bulgarian Section of the Society for Industrial and Applied Mathematics, BGSIAM 2016, Sofia, Bulgaria, 20/12/16. https://doi.org/10.1007/978-3-319-65530-7_17

Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation. / Terziyska, Margarita; Todorov, Yancho; Dobreva, Maria.

Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers. ed. / Michail Todorov; Ivan Georgiev; Krassimir Georgiev; Ivan Georgiev. Springer, 2018. p. 185-201 (Studies in Computational Intelligence, Vol. 728).

Research output: Chapter in Book/Report/Conference proceedingChapter or book articleScientificpeer-review

TY - CHAP

T1 - Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation

AU - Terziyska, Margarita

AU - Todorov, Yancho

AU - Dobreva, Maria

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper the effectiveness of different error metrics for assessment of the capabilities of an advanced fuzzy-neural architecture are studied. The proposed structure combines the potentials of the Intuitionistic Fuzzy Logic with the simplicity of the Neo-Fuzzy Neuron theory for implementation of robust modeling mechanisms, able to capture uncertain variations in the data space. A major concern when evaluating the performance of such kind of models is the selection of appropriate error metrics in order to assess their potential to capture a wide range of system behaviours. Therefore, different error metrics to evaluate the functional properties of a proposed Intuitionistic Neo-fuzzy network are studied and a comparative analysis in modeling of chaotic time series is made.

AB - In this paper the effectiveness of different error metrics for assessment of the capabilities of an advanced fuzzy-neural architecture are studied. The proposed structure combines the potentials of the Intuitionistic Fuzzy Logic with the simplicity of the Neo-Fuzzy Neuron theory for implementation of robust modeling mechanisms, able to capture uncertain variations in the data space. A major concern when evaluating the performance of such kind of models is the selection of appropriate error metrics in order to assess their potential to capture a wide range of system behaviours. Therefore, different error metrics to evaluate the functional properties of a proposed Intuitionistic Neo-fuzzy network are studied and a comparative analysis in modeling of chaotic time series is made.

UR - http://www.scopus.com/inward/record.url?scp=85034248427&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-65530-7_17

DO - 10.1007/978-3-319-65530-7_17

M3 - Chapter or book article

AN - SCOPUS:85034248427

SN - 978-331965529-1

T3 - Studies in Computational Intelligence

SP - 185

EP - 201

BT - Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers

A2 - Todorov, Michail

A2 - Georgiev, Ivan

A2 - Georgiev, Krassimir

A2 - Georgiev, Ivan

PB - Springer

ER -

Terziyska M, Todorov Y, Dobreva M. Efficient Error Based Metrics for Fuzzy-Neural Network Performance Evaluation. In Todorov M, Georgiev I, Georgiev K, Georgiev I, editors, Advanced Computing in Industrial Mathematics - 11th Annual Meeting of the Bulgarian Section of SIAM, Revised Selected Papers. Springer. 2018. p. 185-201. (Studies in Computational Intelligence, Vol. 728). https://doi.org/10.1007/978-3-319-65530-7_17