International Journal of Computational Intelligence Systems

<Previous Article In Issue
Next Article In Issue>
Volume 6, Issue Supplement 1, April 2013, Pages 46 - 60

Quest for Interpretability-Accuracy Trade-off Supported by Fingrams into the Fuzzy Modeling Tool GUAJE

Authors
DavidP. Pancho, JoseM. Alonso, L. Magdalena
Corresponding Author
DavidP. Pancho
Received 19 December 2012, Accepted 7 March 2013, Available Online 29 April 2013.
DOI
10.1080/18756891.2013.818189How to use a DOI?
Keywords
Interpretability-accuracy trade-off, fuzzy modeling, fingrams, GUAJE
Abstract

Understand the behavior of Fuzzy Rule-based Systems (FRBSs) at inference level is a complex task that allows the designer to produce simpler and powerful systems. The fuzzy inference-grams –known as fingrams– establish a novel and mighty tool for understanding the structure and behavior of fuzzy systems. Fingrams represent FRBSs as social networks made of nodes representing fuzzy rules and edges representing the degree of interaction between pairs of rules at inference level (no edge means no significant interaction). We can analyze fingrams obtaining helpful information such as detecting potential conflicts between rules, unused rules and redundant ones. This paper introduces a new module for fingram generation and analysis included in the free software tool GUAJE. This tool aims to design, analyze and evaluate fuzzy systems with good interpretability-accuracy trade-off. In addition, GUAJE includes several intuitive and interactive tutorials to uncover the possibilities it offers. One of them generates and enhances a fuzzy system, analyzing each improvement through the use of fingrams, and lets the user reproduce the illustrative case study described in this paper.

Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

<Previous Article In Issue
Next Article In Issue>
Journal
International Journal of Computational Intelligence Systems
Volume-Issue
6 - Supplement 1
Pages
46 - 60
Publication Date
2013/04/29
ISSN (Online)
1875-6883
ISSN (Print)
1875-6891
DOI
10.1080/18756891.2013.818189How to use a DOI?
Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

risenwbib
TY  - JOUR
AU  - DavidP. Pancho
AU  - JoseM. Alonso
AU  - L. Magdalena
PY  - 2013
DA  - 2013/04/29
TI  - Quest for Interpretability-Accuracy Trade-off Supported by Fingrams into the Fuzzy Modeling Tool GUAJE
JO  - International Journal of Computational Intelligence Systems
SP  - 46
EP  - 60
VL  - 6
IS  - Supplement 1
SN  - 1875-6883
UR  - https://doi.org/10.1080/18756891.2013.818189
DO  - 10.1080/18756891.2013.818189
ID  - Pancho2013
ER  -