multiobjective genetic fuzzy systems - accurate and interpretable fuzzy rule-based classifier design...
Post on 15-Jan-2016
219 views
TRANSCRIPT
![Page 1: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/1.jpg)
Multiobjective Genetic Fuzzy Systems- Accurate and Interpretable Fuzzy Rule-Based
Classifier Design -
Hisao Ishibuchi Osaka Prefecture University, Japan
![Page 2: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/2.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This PresentationAccurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 3: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/3.jpg)
Application Areas of Fuzzy Systems - Fuzzy Control - Fuzzy Clustering - Fuzzy ClassificationControl and clustering are well-known application areas !
Question: Is “fuzzy classification” popular ?
Application Areas of Fuzzy Systems
Web of Science
![Page 4: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/4.jpg)
Application Areas of Fuzzy Systems - Fuzzy Control - Fuzzy Clustering - Fuzzy ClassificationControl and clustering are well-known application areas !
Question: Is “fuzzy classification” popular ?
Application Areas of Fuzzy Systems
Control, Clustering, Classification
Web of Science
![Page 5: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/5.jpg)
Application Areas of Fuzzy SystemsFuzzy Control: Well-Known Application Area
9421 Fuzzy Control Papers
Web of Science
![Page 6: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/6.jpg)
Application Areas of Fuzzy SystemsFuzzy Control: Well-Known Application Area
Takagi-Sugeno Model (1985)
3801ANFIS (1993)
1983
1896
CC Lee: Fuzzy Logic Controller (1990)
Web of Science
![Page 7: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/7.jpg)
Application Areas of Fuzzy SystemsFuzzy Clustering: Well-Known Application Area
2968 Fuzzy Clustering Papers
Web of Science
![Page 8: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/8.jpg)
Application Areas of Fuzzy SystemsFuzzy Clustering: Well-Known Application Area
819
Pal NR, Pal SK
Bezdek JC
Pal NR, Bezdek JC
Web of Science
![Page 9: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/9.jpg)
Application Areas of Fuzzy SystemsFuzzy Classification: Well-Known?
4144 Fuzzy Classification Papers
Web of Science
![Page 10: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/10.jpg)
Application Areas of Fuzzy SystemsFuzzy Classification: Well-Known?
389Dubois D, Prade H
Dubois D, Prade H
Fuzzy Min-Max NN
Fuzzy If-Then Rules
Web of Science
![Page 11: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/11.jpg)
Application Areas of Fuzzy SystemsControl, Clustering, and Classification
9421 (Fuzzy Control)
2968 (Fuzzy Clustering)
4144 (Fuzzy Classification)Web of Science
![Page 12: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/12.jpg)
Application Areas of Fuzzy SystemsControl, Clustering, and Classification
9421 (Fuzzy Control)
2968 (Fuzzy Clustering)
4144 (Fuzzy Classification)
FUZZ-IEEE 1992
IEEE Trans FS 1993FUZZ-IEEE 1992
FUZZ-IEEE 1992
IEEE Trans FS 1993
IEEE Trans FS 1993
Web of Science
![Page 13: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/13.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Maximization- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation Accurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 14: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/14.jpg)
This Presentation: Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Web of Science
![Page 15: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/15.jpg)
This Presentation: Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Fitness Function
w1 Accuracy (S) w2 Complexity (S)
Accuracy Maximization and Complexity Minimization
![Page 16: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/16.jpg)
This Presentation: Accurate and Interpretable Fuzzy Rule-Based Classifier Design
w1 Accuracy (S) w2 Complexity (S)
The number of correctly classified training patterns
The number of selected fuzzy rules
![Page 17: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/17.jpg)
Fuzzy Rules for Classification Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Basic FormIf x1 is small and x2 is small then Class 2
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 18: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/18.jpg)
Fuzzy Rules for Classification Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 19: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/19.jpg)
Fuzzy Rules for Classification Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
If x1 is small and x2 is large then Class 1
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 20: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/20.jpg)
Fuzzy Rules for Classification Accurate and Interpretable Fuzzy Rule-Based Classifier Design
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
If x1 is small and x2 is large then Class 1
. . .
If x1 is large and x2 is large then Class 3
High InterpretabilityEasy to Understand !
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 21: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/21.jpg)
Classification Boundary Accurate and Interpretable Fuzzy Rule-Based Classifier Design
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
If x1 is small and x2 is large then Class 1
. . .
If x1 is large and x2 is large then Class 3
High InterpretabilityEasy to Understand !
![Page 22: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/22.jpg)
Fuzzy Rules for ClassificationBasic form does not always have high accuracy
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
If x1 is small and x2 is large then Class 1
. . .
If x1 is large and x2 is large then Class 3
High InterpretabilityLow Accuracy
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 23: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/23.jpg)
Fuzzy Rules for ClassificationAnother form has a rule weight (certainty)
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
small medium large
largem
edium
small
![Page 24: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/24.jpg)
Classification BoundaryAccurate and Interpretable Fuzzy Rule-Based Classifier Design
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
small medium large
largem
edium
small
![Page 25: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/25.jpg)
Fuzzy Rules with Rule Weights Accurate and Interpretable Fuzzy Rule-Based Classifier Design
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
small medium large
largem
edium
small
H. Ishibuchi et al., Fuzzy Sets and Systems (1992)Web of Science
![Page 26: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/26.jpg)
Fuzzy Rules in This PresentationFuzzy Rules with Rule Weights
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
small medium large
largem
edium
small
Use of Rule Weights: Controversial Issue(1) Rule weight adjustment can be replaced with membership learning.
D. Nauck and R. Kruse, Proc. of FUZZ-IEEE 1998 (1998).Google Scholar
![Page 27: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/27.jpg)
Fuzzy Rules in This PresentationFuzzy Rules with Rule Weights
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
small medium large
largem
edium
small
Use of Rule Weights: Controversial Issue(2) Membership learning can be partially replaced with weight adjustment.
H. Ishibuchi and T. Nakashima, IEEE Trans. FS (2001)Google Scholar
![Page 28: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/28.jpg)
Fuzzy Rules for ClassificationAnother Form: Multiple Consequents
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
Multiple ConsequentsIf x1 is small and x2 is medium
then Class 1 with 0.579, Class 2 with 0.421, Class 3 with 0.000
small medium large
largem
edium
small
![Page 29: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/29.jpg)
Fuzzy Rules for ClassificationAnother Form: Multiple Consequents
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
Basic FormIf x1 is small and x2 is medium
then Class 2
Rule Weight VersionIf x1 is small and x2 is medium
then Class 2 with 0.158
Multiple ConsequentsIf x1 is small and x2 is medium
then Class 1 with 0.579, Class 2 with 0.421, Class 3 with 0.000
small medium large
largem
edium
small
O. Cordon et al., IJAR (2001)Google Scholar
![Page 30: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/30.jpg)
Other Forms of Fuzzy RulesHandling of Classification as Function Approximation
Integer ConsequentIf x1 is small and x1 is large then y = 1
If x1 is large and x2 is large then y = 3
Binary ConsequentIf x1 is small and x1 is large
then (y1, y2, y3) = (1, 0, 0)
If x1 is large and x2 is large
then (y1, y2, y3) = (0, 0, 1)
1.0
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
small medium large
largem
edium
small
![Page 31: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/31.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation Accurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 32: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/32.jpg)
Accuracy ImprovementUse of Fine Fuzzy Partition
S M L
0.0 1.0x1
Class 1 Class 2 Class 3
MS ML
1.00.0
x2
LM
SM
LM
S
![Page 33: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/33.jpg)
Accuracy ImprovementHow to choose an appropriate partition ?
1.0
0.0 1.0
1 2 1.0
0.0 1.0
3 54
1.0
0.0 1.0
6 987 1.0
0.0 1.0
a edb c
1.0
0.0 1.0
f kjg ih1.0
0.0 1.0
l rpm on q
Too Fine Fuzzy Partition ==> Over-Fitting (Poor Generalization Ability)
S M L
0.0 1.0x1
Class 1 Class 2 Class 3
MS ML
1.00.0
x2
LM
SM
LM
S
![Page 34: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/34.jpg)
S M L
0.0 1.0x1
Class 1 Class 2 Class 3
MS ML
1.00.0
x2
LM
SM
LM
S
Accuracy ImprovementHow to choose an appropriate partition ?
One Idea: Use of All Partitions (Multiple Fuzzy Grid Approach)
1.0
0.0 1.0
1 2 1.0
0.0 1.0
3 54
1.0
0.0 1.0
6 987 1.0
0.0 1.0
a edb c
1.0
0.0 1.0
f kjg ih1.0
0.0 1.0
l rpm on q
Ishibuchi et al., Fuzzy Sets and Systems (1992)Web of Science
![Page 35: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/35.jpg)
0.0 1.0x1
Class 1 Class 2 Class 31.00.0
x2
Accuracy ImprovementLearning of Membership Functions
![Page 36: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/36.jpg)
0.0 1.0x1
Class 1 Class 2 Class 31.00.0
x2
Accuracy ImprovementLearning of Membership Functions
Various learning methods such as neuro-fuzzy and genetic-fuzzy methods are available.
D. Nauck and R. Kruse, Fuzzy Sets and Systems (1997).Web of Science
![Page 37: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/37.jpg)
0.0 1.0x1
Class 1 Class 2 Class 31.00.0
x2
Accuracy ImprovementLearning of Membership Functions
Various learning methods such as neuro-fuzzy and genetic-fuzzy methods are available.
Interpretability is degraded.
D. Nauck and R. Kruse, Fuzzy Sets and Systems (1997).Web of Science
![Page 38: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/38.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Accuracy ImprovementUse of Independent Membership Functions
Each fuzzy rule can be generated and adjusted independently from other rules. ==> High Accuracy
Membership functions can be heavily overlapping. ==> Poor Interpretability
![Page 39: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/39.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Accuracy ImprovementUse of Independent Membership Functions
Each fuzzy rule can be generated and adjusted independently from other rules. ==> High Accuracy
Membership functions can be heavily overlapping. ==> Poor Interpretability
S. Abe and M. S. Lan, IEEE Tras. on FS (1995)Web of Science
![Page 40: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/40.jpg)
If x is A then Class 2
Class 1 Class 2 Class 3
A
A: Multi-dimensional Fuzzy Set (Membership Function)
Accuracy ImprovementUse of Multi-Dimensional Membership Functions
![Page 41: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/41.jpg)
If x is A then Class 2
Class 1 Class 2 Class 3
A
A: Multi-dimensional Fuzzy Set (Membership Function)
Accuracy ImprovementUse of Multi-Dimensional Membership Functions
Fuzzy rules are flexibility. ==> High Accuracy
Each membership function is multi-dimensional. ==> Poor Interpretability
![Page 42: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/42.jpg)
If x is A then Class 2
Class 1 Class 2 Class 3
A
A: Multi-dimensional Fuzzy Set (Membership Function)
Accuracy ImprovementUse of Multi-Dimensional Membership Functions
S. Abe et al., IEEE TSMC-C (1998)
S. Abe et al., IEEE TSMC-C (1999)
Web of Science
![Page 43: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/43.jpg)
largesm
all
0.0 1.0x1
Class 1 Class 2 Class 31.00.0
x2
small large
Accuracy ImprovementUse of Tree-Type Fuzzy Partitions
If x1 is small then Class 2.
If x1 is large and x2 is small
then Class 3.
If x1 is large and x2 is large
then Class 1.
![Page 44: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/44.jpg)
x1 is small x1 is large
x2 is small x2 is large
Accuracy ImprovementUse of Tree-Type Fuzzy Partitions
Web of Science
![Page 45: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/45.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation Accurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 46: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/46.jpg)
Difficulty of High-Dimensional ProblemsExponential Increase of Fuzzy Rules
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
. . .
If x1 is large and x2 is large then Class 3
Number of Fuzzy Rules: 2-D Problem: 3x3 3-D Problem: 3x3x3 4-D Problem: 3x3x3x3 5-D Problem: 3x3x3x3x3
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 47: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/47.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Scalability ImprovementUse of Independent Membership Functions
Fuzzy rules are generated in the multi-dimensional space. => No Exponential Increase
![Page 48: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/48.jpg)
If x is A then Class 2
Class 1 Class 2 Class 3
A
A: Multi-dimensional Fuzzy Set (Membership Function)
Scalability ImprovementUse of Multi-Dimensional Membership Functions
Fuzzy rules are generated in the multi-dimensional space. => No Exponential Increase
![Page 49: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/49.jpg)
Scalability ImprovementUse of Tree-Type Fuzzy Partitions
x1 is small x1 is large
x2 is small x2 is large
An appropriate stopping condition prevents the exponential increase in the number of fuzzy rules.
Web of Science
![Page 50: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/50.jpg)
x1 x2 x3 x4 x5
subsystem
subsystemsubsystem
subsystem
Scalability ImprovementHierarchical Fuzzy Systems
subsystem
![Page 51: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/51.jpg)
x1 x2 x3 x4 x5
subsystem subsystem
subsystemsubsystem
subsystem
Scalability ImprovementHierarchical Fuzzy Systems
K. Simojima et al., Fuzzy Sets and Systems (1995)Web of Science
![Page 52: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/52.jpg)
Accuracy & Scalability Improvement ==> Poor Interpretability
- Use of Fine Fuzzy Partition (Accuracy)
- Use of Rule Weights (Accuracy)
- Membership Function Learning (Accuracy)
- Fuzzy Rules with Independent Membership Functions (Accuracy and Scalability)
- Multi-Dimensional Fuzzy Systems (Accuracy and Scalability)
- Tree-Type Fuzzy Partitions (Accuracy and Scalability)
- Hierarchical Fuzzy Systems (Accuracy and Scalability)
![Page 53: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/53.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation Accurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 54: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/54.jpg)
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
Four Rules Nine Rules
Rule Selection (Nine Rules ==> Four Rules)
Complexity MinimizationFuzzy Rule Selection
![Page 55: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/55.jpg)
Four Rules Nine Rules
Rule Selection (Nine Rules ==> Four Rules) The same classification boundaries are generated.
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
Complexity MinimizationFuzzy Rule Selection
![Page 56: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/56.jpg)
Four Rules
Complexity MinimizationFuzzy Rule Selection
Nine Rules
Rule Selection (Nine Rules ==> Four Rules) The same classification boundaries are generated.
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
H. Ishibuchi et al., IEEE Trans. on FS (1995)
Web of Science
![Page 57: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/57.jpg)
Class 1 Class 2 Class 3
Seven Rules Nine Rules
Use of “Don’t Care” Conditions Nine Rules ==> Seven Rules (If x1 is large then Class 3)
Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Complexity MinimizationUse of “Don’t Care” Conditions
![Page 58: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/58.jpg)
Seven Rules Nine Rules
Nine Rules ==> Seven Rules (If x1 is large then Class 3)Slightly different classification boundaries are obtained.
Complexity MinimizationUse of “Don’t Care” Conditions
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
![Page 59: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/59.jpg)
Nine Rules ==> Seven Rules (If x1 is large then Class 3)Slightly different classification boundaries are obtained.
Complexity MinimizationUse of “Don’t Care” Conditions
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
H. Ishibuchi et al., IEEE Trans. on SMC Part B (1999)Google Scholar
![Page 60: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/60.jpg)
The use of “Don’t Care” conditions can be viewed as an scalability improvement method. If x1 is small and x10 is ...
Complexity MinimizationUse of “Don’t Care” Conditions
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
H. Ishibuchi et al., IEEE Trans. on SMC Part B (1999)Google Scholar
![Page 61: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/61.jpg)
The use of “Don’t Care” conditions can be also viewed as input selection for each rule (rule-wise input selection).
Complexity MinimizationUse of “Don’t Care” Conditions
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
H. Ishibuchi et al., IEEE Trans. on SMC Part B (1999)Google Scholar
![Page 62: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/62.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Complexity MinimizationMerging Similar Membership Functions
![Page 63: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/63.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Similar Membership Functions
Complexity MinimizationMerging Similar Membership Functions
![Page 64: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/64.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Similar Membership Functions
==> One Membership Function
Complexity MinimizationMerging Similar Membership Functions
![Page 65: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/65.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Similar Membership Functions
==> One Membership Function
Complexity MinimizationMerging Similar Membership Functions
M. Setnes et al., IEEE Trans. on SMC Part B (1998)
Google Scholar
![Page 66: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/66.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Similar Membership Functions
==> One Membership Function
Complexity MinimizationMerging Similar Membership Functions
- Interpretability is improved.- Accuracy is degraded.
![Page 67: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/67.jpg)
1.00.0 x1
Class 1 Class 2 Class 31.00.0
x2
Similar Membership Functions
==> One Membership Function
Complexity MinimizationMerging Similar Membership Functions
- Interpretability is improved.- Accuracy is degraded.
Interpretability-Accuracy Tradeoff
![Page 68: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/68.jpg)
Complexity MinimizationProjection of Multi-Dimensional Fuzzy Sets
- Interpretability is improved.- Accuracy is degraded.
Interpretability-Accuracy Tradeoff
Aq2
Aq1
0.0 1.00.0
1.0
x1
x2 Aq
If x is Aq then ...
If x1 is Aq1 and x2 is Aq2 then
Change of Fuzzy Rule Form
![Page 69: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/69.jpg)
Interpretability-Accuracy Tradeoff (Accuracy-Simplicity Tradeoff)
Err
or
Sm
all
Lar
ge
ComplexitySimple Complicated
Interpretable
fuzzy system
Accurate
fuzzy system
![Page 70: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/70.jpg)
Accuracy MaximizationMain Research Direction Since the Early 1990s
Err
or
Sm
all
Lar
ge
ComplexitySimple Complicated
Interpretable
fuzzy system
Accurate
fuzzy system
Accuracy Maximization - Fuzzy Neuro Learning - Genetic Fuzzy Optimization
![Page 71: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/71.jpg)
Possible Difficulties
Accuracy Maximization
- Poor Interpretability
- Overfitting to Training Data
![Page 72: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/72.jpg)
Possible Difficulties
Accuracy Maximization
- Poor InterpretabilityPoor Interpretability
- Overfitting to Training Data
![Page 73: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/73.jpg)
Difficulty in Accuracy Maximization
Test data accuracy
Training data accuracy
Accuracy maximization Overfitting
Complexity
Err
or
S*0
![Page 74: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/74.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Complexity
Err
or
y
x
Feasible Area
![Page 75: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/75.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
Complexity
y
x0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Err
or
![Page 76: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/76.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
Complexity
y
x0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Err
or
![Page 77: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/77.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Complexity
y
x
Err
or
![Page 78: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/78.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Complexity
y
x
Err
or
![Page 79: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/79.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Complexity
y
x
Err
or
Test Data Error
Training Data Error
![Page 80: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/80.jpg)
Accuracy-Complexity Tradeoff
Curve Fitting
0 0.2 0.4 0.6 0.8 1
0.2
0.4
0.6
0.8
1
Complexity
y
x
Err
or
Test Data Error
Training Data Error
![Page 81: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/81.jpg)
Possible Difficulties
Accuracy Maximization
- Poor Interpretability
- Overfitting to Training DataOverfitting to Training Data
In the design of fuzzy systems, emphasis should be placed on their linguistic interpretability.
Interpretability maintenance while maximizing the accuracy.
![Page 82: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/82.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Err
or
Sm
al lL
arg
e
ComplexitySimple Complicated
Interpretable
fuzzy system
Accurate
fuzzy system
![Page 83: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/83.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Some Ideas - Aggregated Objective Function: To combine the error
minimization and the complexity minimization into a single scalar fitness function
![Page 84: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/84.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Some Ideas - Aggregated Objective Function: To combine the error
minimization and the complexity minimization into a single scalar fitness function
- Constraint Condition: To use constraint conditions on the position and the shape of membership functions
![Page 85: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/85.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Some Ideas - Aggregated Objective Function: To combine the error
minimization and the complexity minimization into a single scalar fitness function
- Constraint Condition: To use constraint conditions on the position and the shape of membership functions
- Two-Step Fuzzy System Design: 1st Step: Search for accurate and complicated fuzzy rule-based systems. 2nd Step: Simplification of obtained fuzzy rule-based systems.
![Page 86: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/86.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Accuracy Maximization and Complexity Minimization
Google Scholar
![Page 87: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/87.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Basic Idea - Aggregated Objective Function: To combine the error
minimization and the complexity minimization into a single scalar fitness function
- Constraint Condition: To use constraint conditions on the position and the shape of membership functions
- Two-Step Fuzzy System Design: 1st Step: Search for accurate and complicated fuzzy rule-based systems. 2nd Step: Simplification of obtained fuzzy rule-based systems.
![Page 88: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/88.jpg)
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Aggregated Objective Function - To combine the error minimization and the complexity
minimization into a single scalar objective function
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
![Page 89: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/89.jpg)
Compromise between Accuracy and Complexity(Search for a good accuracy-complexity tradeoff)
Aggregated Objective Function - To combine the error minimization and the complexity
minimization into a single scalar objective function
Example: Combination of the average error rate and the number of fuzzy rules
Example of a scalar objective function: Weighted sum
)()()( Complexity2Error1 SfwSfwSf
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
![Page 90: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/90.jpg)
Accuracy and Interpretability MaximizationActive Research Direction Since the Late 1990s
Err
or
Sm
all
Lar
ge
ComplexitySimple Complicated
Interpretable
fuzzy system
Accurate
fuzzy system
Fuzzy systems were automatically generated, trained, and simplified.
![Page 91: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/91.jpg)
Difficulty in Weighted Sum Approach
Sensitivity to the weight vector:
The obtained system strongly depends on the specification of the weight vector.
![Page 92: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/92.jpg)
Complexity
Err
or
Test data accuracy
S*0
Training data accuracy
Minimize w1·Error + w2·Complexity
When the weight for the complexity minimization is large:
A simple system is obtained.
Difficulty in Weighted Sum Approach
![Page 93: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/93.jpg)
Complexity
Err
or
Test data accuracy
S*0
Training data accuracy
Minimize w1·Error + w2·Complexity
When the weight for the error minimization is large:
A complicated system is obtained.
Difficulty in Weighted Sum Approach
![Page 94: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/94.jpg)
Difficulty in Weighted Sum Approach
Complexity
Err
or
Test data accuracy
S*0
Training data accuracy
Minimize w1·Error + w2·Complexity
When the two weights are appropriately specified:
A good system is obtained. But the best complexity is not always found.
![Page 95: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/95.jpg)
Multiobjective Fuzzy System DesignCurrently An Active Research Issue
Basic Idea To search for a number of non-dominated fuzzy systems with
respect to the accuracy maximization and the interpretability maximization (instead of searching for a single fuzzy system).
Aggregation Approach
Multiobjective Approach
)}(),({Minimize ComplexityError SfSf
)()()( Complexity2Error1 SfwSfwSf
![Page 96: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/96.jpg)
Multiobjective Fuzzy System DesignCurrently An Active Research Issue
Basic Idea To search for a number of non-dominated fuzzy systems with
respect to the accuracy maximization and the interpretability maximization (instead of searching for a single fuzzy system).
Aggregation Approach
Multiobjective Approach
)}(),({Minimize ComplexityError SfSf
)()()( Complexity2Error1 SfwSfwSf
Search for Pareto Optimal Fuzzy Rule-Based Systems
![Page 97: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/97.jpg)
Multiobjective Fuzzy System DesignCurrently An Active Research Issue
Err
or
Sm
all
Lar
ge
ComplexitySimple Complicated
Interpretable
fuzzy system
Accurate
fuzzy system
![Page 98: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/98.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This PresentationAccurate and Interpretable Fuzzy Rule-Based Classifier Design
![Page 99: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/99.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page
Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 100: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/100.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page
Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 101: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/101.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 102: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/102.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 103: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/103.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 104: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/104.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 105: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/105.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 106: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/106.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 107: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/107.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 108: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/108.jpg)
Evolutionary Multiobjective Optimization of Fuzzy Rule-Based Systems Bibliography Page Created and Maintained by Marco Cococcioni (Univ. Pisa)
![Page 109: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/109.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 110: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/110.jpg)
Multi-Objective Fuzzy Rule-Based Systems (Fuzzy Rule* OR Fuzzy Rule-Based System*) AND (Multi-Objective OR
Multiobjective OR Two-Objective OR Three-Objective OR Multiple Criteria)
H. Ishibuchi et al., Fuzzy Sets and Systems (1997)
Web of Science
![Page 111: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/111.jpg)
Multi-Objective Fuzzy Rule Selection for Fuzzy Rule-Based Classifier Design
Two Objectives: - The number of correctly classified training patterns - The number of selected fuzzy rules
H. Ishibuchi et al., Fuzzy Sets and Systems (1997)
![Page 112: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/112.jpg)
Multi-Objective Fuzzy Rule-Based Systems (Fuzzy Rule* OR Fuzzy Rule-Based Systems) AND (Multi-Objective OR
Multiobjective OR Two-Objective OR Three-Objective OR Multiple Criteria)
Web of Science
![Page 113: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/113.jpg)
Multi-Objective Fuzzy Rule Selection for Fuzzy Rule-Based Classifier Design
Two Objectives: - The number of correctly classified training patterns - The number of selected fuzzy rules
H. Ishibuchi et al., Fuzzy Sets and Systems (1997)
H. Ishibuchi et al., Information Science (2001)
Three Objectives: - The number of correctly classified training patterns - The number of selected fuzzy rules - Total number of antecedent conditions (Total rule length)
![Page 114: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/114.jpg)
Multi-Objective Fuzzy Rule-Based Systems (Fuzzy Rule* OR Fuzzy Rule-Based Systems) AND (Multi-Objective OR
Multiobjective OR Two-Objective OR Three-Objective OR Multiple Criteria)
Web of Science
![Page 115: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/115.jpg)
Multi-Objective RBF Function Design for Function Approximation Problems
![Page 116: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/116.jpg)
Multi-Objective RBF Function Design for Function Approximation Problems
Multi-Objective Neural Network Design and Learning
![Page 117: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/117.jpg)
Multiobjective Neural Networks
0Complexity
Err
or
![Page 118: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/118.jpg)
Multiobjective Decision Trees (GP)
0Complexity
Err
or
![Page 119: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/119.jpg)
Multi-Objective Fuzzy Rule-Based Systems (Fuzzy Rule* OR Fuzzy Rule-Based Systems) AND (Multi-Objective OR
Multiobjective OR Two-Objective OR Three-Objective OR Multiple Criteria)
Web of Science
![Page 120: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/120.jpg)
Paper Title: Industrial Applications of Fuzzy-Logic at General-Electric
Web of Science
![Page 121: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/121.jpg)
Paper Title: Industrial Applications of Fuzzy-Logic at General-Electric
Web of Science
![Page 122: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/122.jpg)
Recent Publications on Multi-Objective Fuzzy System Design
Web of Science
![Page 123: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/123.jpg)
Recent Publications on Multi-Objective Fuzzy System Design
Web of Science
![Page 124: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/124.jpg)
Paper Title: A Multiobjective Evolutionary Approach to Concurrently Learn Rule and Data Bases of Linguistic Fuzzy-Rule-Based Systems
Web of Science
![Page 125: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/125.jpg)
Paper Title: A Multiobjective Evolutionary Approach to Concurrently Learn Rule and Data Bases of Linguistic Fuzzy-Rule-Based Systems
1. Univ Granada 2. Univ Pisa
Web of Science
![Page 126: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/126.jpg)
Paper Title: A Multiobjective Evolutionary Approach to Concurrently Learn Rule and Data Bases of Linguistic Fuzzy-Rule-Based Systems
1. Univ Granada 2. Univ Pisa
Web of Science
![Page 127: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/127.jpg)
Paper Title: Learning concurrently partition granularities and rule bases of Mamdani fuzzy systems in a multi-objective evolutionary framework
Web of Science
![Page 128: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/128.jpg)
Paper Title: Learning concurrently partition granularities and rule bases of Mamdani fuzzy systems in a multi-objective evolutionary framework
1. Univ Pisa
Web of Science
![Page 129: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/129.jpg)
Paper Title: Learning concurrently partition granularities and rule bases of Mamdani fuzzy systems in a multi-objective evolutionary framework
1. Univ Pisa
Web of Science
![Page 130: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/130.jpg)
Multi-Objective Fuzzy System Design ResearchActive Geographical RegionsGoogle Maps
![Page 131: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/131.jpg)
Multi-Objective Fuzzy System Design ResearchActive Geographical RegionsGoogle Maps
![Page 132: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/132.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 133: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/133.jpg)
Complexity0
Err
orMany non-dominated fuzzy systems can be obtained along the tradeoff surface by a single run of an EMO algorithm.
EMO: Evolutionary Multi-Objective Optimization
Results of Multi-Objective SearchNon-Dominated Fuzzy Rule-Based Systems
A Single Fuzzy Rule-Based System
![Page 134: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/134.jpg)
Accuracy-Complexity Tradeofffor Training Data and Test Data
Complexity
Test data
S*0
Training data
Err
or
The obtained non-dominated fuzzy systems show the tradeoff between the complexity and the training data accuracy (not the tradeoff between the complexity and the test data accuracy).
![Page 135: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/135.jpg)
Accuracy-Complexity Tradeofffor Training Data and Test Data
Complexity
Test data
S*0
Training data
Err
or
The obtained non-dominated fuzzy systems show the tradeoff between the complexity and the training data accuracy (not the tradeoff between the complexity and the test data accuracy).
- Tradeoff for test data accuracy should be examined.- This can be done since we have many fuzzy systems.
![Page 136: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/136.jpg)
Example: Obtained Rule Sets (Heart C)
Training data accuracy Testing data accuracy
Number of rules
Err
or r
ate
(%)
0 2 4 6 8 10 12 1410
20
30
40
50
60
Number of rulesE
rror
rat
e (%
)0 2 4 6 8 10 12 14
30
40
50
60
70
80
Obtained rule sets help us to find the optimal complexity of fuzzy systems. (Rule sets with six, seven and eight rules may be good)
![Page 137: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/137.jpg)
A rule set with eight fuzzy rules
Some human users may prefer simpler rule sets.
A Rule Set with High-Generalization Ability
Number of rules
Err
or r
ate
(%)
0 2 4 6 8 10 12 1410
20
30
40
50
60
Number of rules
Err
or r
ate
(%)
Class 1(0.23)
Class 1(0.81)
Consequent
Class 2(0.63)
Class 2(0.20)
Class 2(1.00)
Class 3(0.35)
Class 3(0.24)
x1 x3 x4 x6 x7 x8 x10 x11 x12
R1
R2
R3
R4
R5
R6
R7
R8
Class 1(0.46)
![Page 138: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/138.jpg)
A very simple rule set with only two fuzzy rules
A Rule Set with High Interpretability
Number of rules
Err
or r
ate
(%)
0 2 4 6 8 10 12 1410
20
30
40
50
60
Number of rules
Err
or r
ate
(%) 10x
1R
2R
Consequent
Class 1(0.26)
Class 2(1.00)
11x
DC
DC
![Page 139: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/139.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 140: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/140.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 141: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/141.jpg)
1. Large Search Space: Difficulty in Search
The search space exponentially increases with the number of attributes (i.e., with the dimensionality of the pattern space).
Why is the fuzzy system design difficult?
![Page 142: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/142.jpg)
Number of Fuzzy Rules
Basic FormIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
. . .
If x1 is large and x2 is large then Class 3
Number of Fuzzy Rules: 2-D Problem: 3x3 3-D Problem: 3x3x3 4-D Problem: 3x3x3x3 5-D Problem: 3x3x3x3x3
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 143: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/143.jpg)
Number of Fuzzy Rules
Use of Don’t CareIf x1 is small and x2 is small then Class 2
If x1 is small and x2 is medium then Class 2
. . .
If x1 is large and x2 is don’t care then Class 3
Number of Fuzzy Rules: 2-D Problem: (3+1)x(3+1) 3-D Problem: (3+1)3
4-D Problem: (3+1)4
5-D Problem: (3+1)5
1.0
small medium large
0.0 1.0x1
Class 1 Class 2 Class 3
0.0x
2
largem
edium
small
![Page 144: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/144.jpg)
Search Space Size: Large
Example: Classification problem with 50 attributes and 3 linguistic values for each attribute
Number of Fuzzy Rules and Number of Rule Sets
1.0
small
medium large
0.0 1.0x1
0.0x
2
large
mediu
msm
all
Class 1 Class 2 Class 3The total number of fuzzy rules (i.e., antecedent condition combinations):
(3+1) x ... x (3+1) = 450 = 2100
The total number of fuzzy rule sets with 20 rules (i.e., combinations of 20 fuzzy rules):
NC20 ~ 22000 where N = 2100
![Page 145: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/145.jpg)
Search Space Size: Large
Example: Classification problem with 50 attributes and 1-7 fuzzy partition for each attribute
Number of Fuzzy Rules and Number of Rule Sets
The total number of fuzzy rules (i.e., antecedent condition combinations):
(1+2+ ... 7) x ... = 2850 > 2400
The total number of fuzzy rule sets with 20 rules (i.e., combinations of 20 fuzzy rules):
NC20 ~ 28000 where N > 2400
1.0
0.0 1.0
1 2 1.0
0.0 1.0
3 54
1.0
0.0 1.0
6 987 1.0
0.0 1.0
a edb c
1.0
0.0 1.0
f kjg ih1.0
0.0 1.0
l rpm on q
![Page 146: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/146.jpg)
1. Large Search Space: Difficulty in Search
The search space exponentially increases with the number of attributes (i.e., with the dimensionality of the pattern space). It is likely that the entire tradeoff curve can not be covered well by the obtained non-dominated fuzzy rule-based systems.
Why is the fuzzy system design difficult?
Complexity0
Err
or
0 Complexity
Err
or
Poor Diversity Poor Convergence
![Page 147: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/147.jpg)
1. Large Search Space: Difficulty in Search
The search space exponentially increases with the number of attributes (i.e., with the dimensionality of the pattern space). It is likely that the entire tradeoff curve can not be covered well by the obtained non-dominated fuzzy rule-based systems.
Why is the fuzzy system design difficult?
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy.
![Page 148: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/148.jpg)
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy. This means that the fitness function improvement does not always lead to better fuzzy rule-based classifiers (when the training data accuracy is used in the fitness function) .
1. Large Search Space: Difficulty in Search
The search space exponentially increases with the number of attributes (i.e., with the dimensionality of the pattern space). It is likely that the entire tradeoff curve can not be covered well by the obtained non-dominated fuzzy rule-based systems.
Why is the fuzzy system design difficult?
![Page 149: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/149.jpg)
Why is the fuzzy system design difficult?
Complexity0
Err
or
Poor Diversity
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy.
Test Data Accuracy
Training Data Accuracy
Training Data Accuracy Improvement
![Page 150: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/150.jpg)
Why is the fuzzy system design difficult?
Complexity0
Err
or
Poor Diversity Diversity Improvement
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy.
Test Data Accuracy
Complexity0
Err
or
Test Data Accuracy
Training Data Accuracy Training Data Accuracy
![Page 151: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/151.jpg)
Why is the fuzzy system design difficult?
Complexity0
Err
or
Poor Convergence
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy.
Test Data Accuracy
Training Data Accuracy
Training Data Accuracy Improvement
![Page 152: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/152.jpg)
Why is the fuzzy system design difficult?
Complexity0
Err
or
Poor Convergence Convergence Improvement
2. Possibility of Over-Fitting: Difficulty in Learning
The improvement in the training data accuracy does not always mean the improvement in the test data accuracy.
Test Data Accuracy
Complexity0
Err
or
Test Data Accuracy
Training Data AccuracyTraining Data Accuracy
![Page 153: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/153.jpg)
Recent Studies: Improvement in training data accuracy leads to Improvement in test data accuracy.
Poor Diversity Diversity ImprovementComplexity0
Err
or
Test Data Accuracy
Complexity0
Err
or
Test Data Accuracy
Training Data Accuracy Training Data Accuracy
Gacto MJ, Alcala R, Herrera F (2009) Adaptation and application of multi-objective evolutionary algorithms for rule reduction and parameter tuning of fuzzy rule-based systems, Soft Computing 13 (5): 419-436
Ishibuchi H, Nakashima Y, Nojima Y (2009) Performance evaluation of evolutionary multiobjective optimization algorithms for multiobjective fuzzy genetics-based machine learning, Proc. of FUZZ-IEEE 2009.
![Page 154: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/154.jpg)
Our Experimental Results
MoFGBML Algorithm (Framework: NSGA-II) Multi-Objective Fuzzy Genetics-Based Machine Learning
Ishibuchi & Nojima, IJAR 2007
NSGA-II Basic Setting - Population size: 200 individuals - Termination Condition: 2000 generations - Multiple Fuzzy Partitions: Granularities 1-5
Three Variants of MoFGBML Setting - Diversity Improvement Method (Mating, EJOR 2008) - Termination Condition: 20000 generations - Multiple Fuzzy Partitions: Granularities 7
Google Scholar
![Page 155: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/155.jpg)
Experimental Results (Glass)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
NSGA-II Basic Setting
Training Data Accuracy
Improvement
![Page 156: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/156.jpg)
Experimental Results (Glass)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020 5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
NSGA-II Basic Setting NSGA-II with Mating
![Page 157: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/157.jpg)
Experimental Results (Glass)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020 5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
NSGA-II Basic Setting NSGA-II with Mating
![Page 158: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/158.jpg)
Experimental Results (Glass)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020 5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
NSGA-II 20,000 Generations NSGA-II with Mating
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
2,000 generations
20,000 generations
![Page 159: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/159.jpg)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 30205 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
Experimental Results (Glass)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
NSGA-II Granularities 1-7 NSGA-II with Mating
1-5 granularities
1-7 granularities
![Page 160: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/160.jpg)
Our Experimental ResultsSimple Changes of Objectives (GEFS 2010, Spain)
Original Formulation f1(S): Error Rate (%) f2(S): Number of Fuzzy Rules
Simple Modification g1(S) = f1(S) - f2(S) g2(S) = f2(S) + f1(S)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
f1(S)
f2(S)
g2(S)
g2(S)
![Page 161: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/161.jpg)
Our Experimental ResultsSimple Changes of Objectives (GEFS 2010, Spain)
5 10 15 20 25 30 350.10
0.15
0.20
0.25
0.30
0.35
Number of fuzzy rules
Err
or r
ate
on tr
aini
ng d
ata
(%) 35
30
25
20
15
10 5 10 15 20 25 30 35
0 0.01 0.1
5 10 15 20 25 30 350.14
0.16
0.18
0.20
0.22
0.24
0.26 0 0.01 0.1
Number of fuzzy rules
Err
or r
ate
on tr
aini
ng d
ata
(%) 26
24
22
20
18
14 5 10 15 20 25 30 35
16
(a) Glass data. (b) Diabetes data.
Original Problem Original Problem
![Page 162: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/162.jpg)
Our Experimental ResultsSimple Changes of Objectives (GEFS 2010, Spain)
Original Formulation f1(S): Error Rate (%) f2(S): Number of Fuzzy Rules
Simple Modification g1(S) = f1(S) - f2(S) g2(S) = f2(S) + f1(S)
5 10 15 20 25 300.1
0.2
0.3
0.4
0.5
0.6Test Data
Training Data
Number of fuzzy rules
Err
or r
ate
(%)
60
50
40
30
20
10 5 10 15 25 3020
f1(S)
f2(S)
Four-Objective g1(S) = f1(S) - f2(S) g2(S) = f1(S) + f2(S) g3(S) = f2(S) - f1(S) g4(S) = f2(S) + f1(S)
![Page 163: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/163.jpg)
Our Experimental ResultsFour-Objective Formulation (GEFS 2010, Spain)
(a) Glass data. (b) Diabetes data.
Original Problem Original Problem
5 10 15 20 25 30 350.10
0.15
0.20
0.25
0.30
0.35
Number of fuzzy rules
Err
or r
ate
on tr
aini
ng d
ata
(%) 35
30
25
20
15
10 5 10 15 20 25 30 35
0 0.01 0.1
5 10 15 20 25 30 350.14
0.16
0.18
0.20
0.22
0.24
0.26 0 0.01 0.1
Number of fuzzy rules
Err
or r
ate
on tr
aini
ng d
ata
(%) 26
24
22
20
18
14 5 10 15 20 25 30 35
16
![Page 164: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/164.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 165: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/165.jpg)
Handling of Interpretabilityin Our Former Studies[1] H. Ishibuchi et al. (1995) Selecting fuzzy if-then rules for
classification problems using genetic algorithms, IEEE TFS.ISI Web of Knowledge Citations: 273 times
[2] H. Ishibuchi et al. (1997) Single-objective and two-objective genetic algorithms for selecting linguistic rules for pattern classification problems. Fuzzy Sets & Systems.ISI Web of Knowledge Citations: 107 times
[3] H. Ishibuchi et al. (2001) Three-objective genetics-based machine learning for linguistic rule extraction. Information Sciences.ISI Web of Knowledge Citations: 94 times
Interpretability Maximization = Complexity Minimization - Minimization of the number of fuzzy rules - Minimization of the number of antecedent conditions
![Page 166: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/166.jpg)
Interpretability of Fuzzy Systems
Ishibuchi et al. (1995, 1997, 2001)Interpretability Maximization = Complexity Minimization - Minimization of the number of fuzzy rules - Minimization of the number of antecedent conditions
![Page 167: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/167.jpg)
Interpretability of Fuzzy Systems
Ishibuchi et al. (1995, 1997, 2001)Interpretability Maximization = Complexity Minimization - Minimization of the number of fuzzy rules - Minimization of the number of antecedent conditions
Many other factors are related to the interpretability
![Page 168: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/168.jpg)
Interpretability of Fuzzy Systems
Ishibuchi et al. (1995, 1997, 2001)Interpretability Maximization = Complexity Minimization - Minimization of the number of fuzzy rules - Minimization of the number of antecedent conditions
Many other factors are related to the interpretability
Special Sessions and Many Related Papers - IFSA 2009 Conference - ISDA 2009 Conference (4 Papers with Interpretability in Their Titles)
- FUZZ-IEEE 2010 Conference
![Page 169: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/169.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 170: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/170.jpg)
1.0small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
Another Issue in InterpretabilityExplanation of Classification Results
Explanation AbilityThe ability of fuzzy rule-based systems to explain why a new pattern is classified as a particular class.
Example: Classification of a pattern : xA = (0.05, 0.05)
xA
![Page 171: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/171.jpg)
(2) Rule Set 2: Four Rules
Comparison between Rule Sets 1 and 2
(1) Rule Set 1: Nine Rules
Classification of : xA = (0.05, 0.05)
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xA xA
![Page 172: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/172.jpg)
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
R1: If x1 is small and x2 is small then Class 2
Comparison in Explanation CapabilityResponsible Rules for Classification
R5: If x1 is medium and x2 is medium then Class 2
xA xA
![Page 173: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/173.jpg)
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
R1: If x1 is small and x2 is small then Class 2
Comparison in Explanation CapabilityResponsible Rules for Classification
R5: If x1 is medium and x2 is medium then Class 2
xA xA
R1 seems to be a better explanation for the classification of xA.
![Page 174: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/174.jpg)
(2) Rule Set 2: Four Rules
Comparison between Rule Sets 1 and 2
(1) Rule Set 1: Nine Rules
Rule Set 1 seems to have higher explanation ability while Rule Set 2 is simpler than Rule Set 1.
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xA xA
![Page 175: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/175.jpg)
Comparison between Rule Sets 1 and 4
(4) Rule Set 4: Three Rules
(1) Rule Set 1: Nine Rules
Classification of : xA = (0.05, 0.05)
Class 1 Class 2 Class 3 Class 1 Class 2 Class 3
small or medium large
largem
edium or sm
all
0.0 1.0x1
1.00.0
x2
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xA xA
![Page 176: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/176.jpg)
small or medium large
largem
edium or sm
all
0.0 1.0x1
1.00.0
x2
R1234: If x1 is small or medium and x2 is small or medium then Class 2
R1: If x1 is small and x2 is small then Class 2
Comparison in Explanation CapabilityResponsible Rules for Classification
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 Class 1 Class 2 Class 3
xA xA
![Page 177: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/177.jpg)
small or medium large
largem
edium or sm
all
0.0 1.0x1
1.00.0
x2
R1234: If x1 is small or medium and x2 is small or medium then Class 2
R1: If x1 is small and x2 is small then Class 2
Comparison in Explanation CapabilityResponsible Rules for Classification
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 Class 1 Class 2 Class 3
xA xA
R1 seems to be a better explanation for the classification of xA.
![Page 178: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/178.jpg)
Comparison between Rule Sets (1) and (4)
(4) Rule Set 4: Three Rules
(1) Rule Set 1: Nine Rules
Rule Set 1 seems to have higher explanation ability while Rule Set 4 is simpler than Rule Set 1.
Class 1 Class 2 Class 3 Class 1 Class 2 Class 3
small or medium large
largem
edium or sm
all
0.0 1.0x1
1.00.0
x2
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xA xA
![Page 179: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/179.jpg)
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3
Classification CapabilityAnother Example xB= (0.95, 0.50)
Classification of : xB = (0.95, 0.50)
xB
![Page 180: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/180.jpg)
Comparison between Rule Sets 1 and 3
(3) Rule Set 3: Seven Rules
(1) Rule Set 1: Nine Rules
Classification of : xB = (0.95, 0.50)
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xB xB
![Page 181: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/181.jpg)
Comparison in Explanation CapabilityResponsible Rules for Classification
R8: If x1 is large and x2 is medium then Class 3
R789: If x1 is large then Class 3
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xB xB
![Page 182: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/182.jpg)
Comparison in Explanation CapabilityResponsible Rules for Classification
R8: If x1 is large and x2 is medium then Class 3
R789: If x1 is large then Class 3
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xB xB
Which is a better explanation for the classification of xB between R8 and R789 ?
![Page 183: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/183.jpg)
Comparison in Explanation CapabilityResponsible Rules for Classification
R8: If x1 is large and x2 is medium then Class 3
R789: If x1 is large then Class 3
1.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
Class 1 Class 2 Class 3 Class 1 Class 2 Class 31.0
small medium large
0.0 1.0x1
0.0x
2
largem
edium
small
xB xB
Which is a better explanation for the classification of xB between R8 and R789 ? It is a very difficult question for me to answer.
![Page 184: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/184.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Contents of This Presentation
![Page 185: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/185.jpg)
We have a lot of different types of classification problems where fuzzy rule-based classifiers have not been well-utilized and have a large potential usefulness:
1. Imbalanced Data2. Semi-Supervised Learning3. Active Learning4. On-Line Learning5. . . .6. . . .
Fuzzy Classifiers on Various Problems
![Page 186: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/186.jpg)
Fuzzy Classifiers on Various Problems
IFSA 2009 Invited Talk
![Page 187: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/187.jpg)
Fuzzy Classifiers on Various Problems
IFSA 2009 Invited Talk
![Page 188: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/188.jpg)
Fuzzy Classifiers on Various Problems
IFSA 2009 Invited Talk
Machine Learning
Basic Advanced
Fuzzy
![Page 189: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/189.jpg)
Fuzzy Classifiers on Various Problems
ISDA 2009 Invited Talk by Hisao Ishibuchi
Machine Learning
Basic Advanced
Fuzzy
MoFuzzyMoML
![Page 190: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/190.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Conclusions
Yes !
![Page 191: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/191.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Conclusions
It is not always high.
![Page 192: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/192.jpg)
1. Introduction to Fuzzy Rule-Based Classification - Is Fuzzy Rule-Based Classification a Popular Research Area?
2. Fuzzy Rule-Based Classifier Design- Accuracy Improvement- Scalability to High-Dimensional Problems- Complexity Minimization
3. Multiobjective Fuzzy Rule-Based Classifier Design- Formulation of Multi-objective Problems- Accuracy-Complexity Tradeoff Analysis- Maximization of Generalization Ability
4. Current Hot Issues and Future Research Directions- Search Ability of EMO for Fuzzy System Design - Definition of Interpretability of Fuzzy Systems- Explanation Ability of Fuzzy Rule-Based Systems- Various Classification Problems: Imbalanced, Online, ...
Conclusions
We have still a lot of interesting research issues.
![Page 193: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/193.jpg)
Appendix: Comparison of the Two Approaches
Two-objective maximization problem
EMO Approach Weighted Sum ApproachExperimental results of a single run of each approach
f1: Total profit from knapsack 1
f 2: T
otal
pro
fit f
rom
kna
psac
k 2 2000th generation
50th generation20th generation
Pareto front
16000 17000 18000 19000 2000016000
17000
18000
19000
20000
21000
f1: Total profit from knapsack 1
f 2: T
otal
pro
fit f
rom
kna
psac
k 2 2000th generation
50th generation20th generation
Pareto front
16000 17000 18000 19000 2000016000
17000
18000
19000
20000
21000
![Page 194: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/194.jpg)
Appendix: Two-Dimensional Antecedent Fuzzy Sets
Aq
Aq2
Aq1
0.0 1.00.0
1.0
x1
x2
Aq2
Aq1
0.0 1.00.0
1.0
x1
x2 Aq
(a) A two-dimensional fuzzy vector. (b) An ellipsoidal antecedent fuzzy set.
![Page 195: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/195.jpg)
Appendix: Interval Rules vs Fuzzy Rules
(a) Interval Rules (b) Fuzzy Rules
S L
Class 1 Class 2
0.0 x1 1.0
0.0
x2
1.0
M
SL
M
S L
SL
Class 1 Class 2
0.0 x1 1.0
0.0
x2
1.0
M
M
![Page 196: Multiobjective Genetic Fuzzy Systems - Accurate and Interpretable Fuzzy Rule-Based Classifier Design - Hisao Ishibuchi Osaka Prefecture University, Japan](https://reader036.vdocument.in/reader036/viewer/2022081603/56649d455503460f94a22648/html5/thumbnails/196.jpg)
Appendix: Interval Rules vs Fuzzy Rules
(a) Interval Rules (b) Fuzzy Rules
S L
Class 1 Class 2
0.0 x1 1.0
0.0
x2
1.0
M
SL
M
S L
SL
Class 1 Class 2
0.0 x1 1.0
0.0
x2
1.0
M
M