New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems

Krisztian Balazs, L. Kóczy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

It is well known that beyond the fact that fuzzy systems have favorable modeling capabilities from the viewpoint of accuracy, they also have outstanding inherent interpretability possibilities, which is a rather unique property among modeling architectures and which is a strong motivation for their research and application. This paper focuses on both mentioned property types and proposes a new technique for adjusting between accuracy and interpretability in modeling systems where fuzzy rule based architectures together with evolutionary algorithms are used for knowledge extraction. First, an inconsistency problem of conventional interpretable fuzzy systems is resolved. Then, a new search space narrowing technique for evolutionary algorithms is proposed, which can be applied for constructing interpretable fuzzy rule bases. Finally, the favorable properties of this new approach will be verified experimentally by carrying out simulation runs.

Original languageEnglish
Title of host publicationCINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings
Pages323-328
Number of pages6
DOIs
Publication statusPublished - 2012
Event13th IEEE International Symposium on Computational Intelligence and Informatics, CINTI 2012 - Budapest, Hungary
Duration: Nov 20 2012Nov 22 2012

Other

Other13th IEEE International Symposium on Computational Intelligence and Informatics, CINTI 2012
CountryHungary
CityBudapest
Period11/20/1211/22/12

Fingerprint

Fuzzy rules
Fuzzy systems
Evolutionary algorithms

Keywords

  • Fuzzy systems
  • Interpretability
  • Knowledge extraction

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Information Systems

Cite this

Balazs, K., & Kóczy, L. (2012). New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems. In CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings (pp. 323-328). [6496783] https://doi.org/10.1109/CINTI.2012.6496783

New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems. / Balazs, Krisztian; Kóczy, L.

CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings. 2012. p. 323-328 6496783.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Balazs, K & Kóczy, L 2012, New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems. in CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings., 6496783, pp. 323-328, 13th IEEE International Symposium on Computational Intelligence and Informatics, CINTI 2012, Budapest, Hungary, 11/20/12. https://doi.org/10.1109/CINTI.2012.6496783
Balazs K, Kóczy L. New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems. In CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings. 2012. p. 323-328. 6496783 https://doi.org/10.1109/CINTI.2012.6496783
Balazs, Krisztian ; Kóczy, L. / New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems. CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings. 2012. pp. 323-328
@inproceedings{b8364248592d46ada8844e82bd02ac57,
title = "New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems",
abstract = "It is well known that beyond the fact that fuzzy systems have favorable modeling capabilities from the viewpoint of accuracy, they also have outstanding inherent interpretability possibilities, which is a rather unique property among modeling architectures and which is a strong motivation for their research and application. This paper focuses on both mentioned property types and proposes a new technique for adjusting between accuracy and interpretability in modeling systems where fuzzy rule based architectures together with evolutionary algorithms are used for knowledge extraction. First, an inconsistency problem of conventional interpretable fuzzy systems is resolved. Then, a new search space narrowing technique for evolutionary algorithms is proposed, which can be applied for constructing interpretable fuzzy rule bases. Finally, the favorable properties of this new approach will be verified experimentally by carrying out simulation runs.",
keywords = "Fuzzy systems, Interpretability, Knowledge extraction",
author = "Krisztian Balazs and L. K{\'o}czy",
year = "2012",
doi = "10.1109/CINTI.2012.6496783",
language = "English",
isbn = "9781467352062",
pages = "323--328",
booktitle = "CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings",

}

TY - GEN

T1 - New parameterizable search space narrowing technique for adjusting between accuracy and interpretability in fuzzy systems

AU - Balazs, Krisztian

AU - Kóczy, L.

PY - 2012

Y1 - 2012

N2 - It is well known that beyond the fact that fuzzy systems have favorable modeling capabilities from the viewpoint of accuracy, they also have outstanding inherent interpretability possibilities, which is a rather unique property among modeling architectures and which is a strong motivation for their research and application. This paper focuses on both mentioned property types and proposes a new technique for adjusting between accuracy and interpretability in modeling systems where fuzzy rule based architectures together with evolutionary algorithms are used for knowledge extraction. First, an inconsistency problem of conventional interpretable fuzzy systems is resolved. Then, a new search space narrowing technique for evolutionary algorithms is proposed, which can be applied for constructing interpretable fuzzy rule bases. Finally, the favorable properties of this new approach will be verified experimentally by carrying out simulation runs.

AB - It is well known that beyond the fact that fuzzy systems have favorable modeling capabilities from the viewpoint of accuracy, they also have outstanding inherent interpretability possibilities, which is a rather unique property among modeling architectures and which is a strong motivation for their research and application. This paper focuses on both mentioned property types and proposes a new technique for adjusting between accuracy and interpretability in modeling systems where fuzzy rule based architectures together with evolutionary algorithms are used for knowledge extraction. First, an inconsistency problem of conventional interpretable fuzzy systems is resolved. Then, a new search space narrowing technique for evolutionary algorithms is proposed, which can be applied for constructing interpretable fuzzy rule bases. Finally, the favorable properties of this new approach will be verified experimentally by carrying out simulation runs.

KW - Fuzzy systems

KW - Interpretability

KW - Knowledge extraction

UR - http://www.scopus.com/inward/record.url?scp=84876907509&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876907509&partnerID=8YFLogxK

U2 - 10.1109/CINTI.2012.6496783

DO - 10.1109/CINTI.2012.6496783

M3 - Conference contribution

AN - SCOPUS:84876907509

SN - 9781467352062

SP - 323

EP - 328

BT - CINTI 2012 - 13th IEEE International Symposium on Computational Intelligence and Informatics, Proceedings

ER -