Analysis of interpretability-accuracy tradeoff of fuzzy systems by multiobjective fuzzy genetics-based machine learning

Ishibuchi, Hisao; Nojima, Yusuke

doi:10.1016/j.ijar.2006.01.004

Cited by 344 publications

(214 citation statements)

References 25 publications

Supporting

Mentioning

204

Contrasting

Unclassified

Order By: Relevance

“…Here, a complete FPT is specified by the expert, including the structure and the type of aggregation functions. Only the parameters of this model (i.e., the parameters of the aggregation functions and the parameters of the underlying fuzzy sets (2)) are calibrated in a data-driven way. To this end, the calibration procedure proposed in [20] has been used, which is based on evolutionary optimization techniques.…”

Section: Hybrid Modelingmentioning

confidence: 99%

See 1 more Smart Citation

Fuzzy pattern trees as an alternative to rule-based fuzzy systems: Knowledge-driven, data-driven and hybrid modeling of color yield in polyester dyeing

Nasiri

Fober

Senge

et al. 2013

2013 Joint IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS)

View full text Add to dashboard Cite

This paper advocates a novel approach to fuzzy systems modeling called fuzzy pattern trees. This approach is largely motivated by alleged disadvantages of rule-based system architectures that still dominate the field. Due to its hierarchical, modular structure and the use of different types of (nonlinear) aggregation operators, a fuzzy pattern tree has the ability to represent functional dependencies in a more flexible and more compact way, thereby offering a reasonable balance between accuracy and model transparency. We evaluate this new model class in the context of a concrete case study, namely the modeling of color yield in polyester high temperature dyeing as a function of disperse dyes concentration, temperature and time. To this end, we compare three possibilities for model construction: purely knowledge-driven, purely data-driven and a hybrid approach combining these two. Our results show that, in comparison to conventional fuzzy modeling using Mamdani rules, fuzzy pattern trees are not only more accurate but also more compact and therefore more easily interpretable, regardless of whether the models are constructed in a knowledge-driven, data-driven or hybrid manner. Moreover, we show that a hybrid modeling approach can outperform a purely data-driven and a purely knowledge-driven approach if expert knowledge and model calibration are combined in a suitable way.

show abstract

Section: Hybrid Modelingmentioning

confidence: 99%

“…Another problem that may hamper interpretability concerns the complexity of models consisting of a potentially large number of interacting pieces, for example rules in a rule-based system. Since accurate models typically require a certain level of complexity, accuracy and understandability are to some extent conflicting goals [1,2].…”

Section: Introductionmentioning

confidence: 99%

Fuzzy pattern trees as an alternative to rule-based fuzzy systems: Knowledge-driven, data-driven and hybrid modeling of color yield in polyester dyeing

Nasiri

Fober

Senge

et al. 2013

2013 Joint IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS)

View full text Add to dashboard Cite

show abstract

“…Weights z q j of this type were studied in [48]. In [60], weights based on ROC analysis are advocated.…”

Section: The Fuzzy Modelmentioning

confidence: 99%

“…Such an approach has also been adopted for learning weights z q j of fuzzy rules [47]. In [48,49], genetic search-based multi-objective optimization was applied to design a fuzzy rule-based system. The task was to maximize f 1 (S), minimize f 2 (S), and minimize f 3 (S), where S is a set of fuzzy rules, f 1 (S) stands for correctly classified training samples, f 2 (S) is the number of fuzzy rules in S, and f 3 (S) is the total number of antecedent conditions in S. Thus, f 3 (S) can be considered as the total rule length.…”

mentioning

confidence: 99%

A general framework for designing a fuzzy rule-based classifier

et al. 2010

View full text Add to dashboard Cite

This is an accepted version of a paper published in Knowledge and Information Systems. This paper has been peer-reviewed but does not include the final publisher proofcorrections or journal pagination.Citation for the published paper: verikas, a., guzaitis, j., gelzinis, a., bacauskiene, m. Abstract This paper presents a general framework for designing a fuzzy rule-based classifier. Structure and parameters of the classifier are evolved through a two-stage genetic search. To reduce the search space, the classifier structure is constrained by a tree created using the evolving SOM tree algorithm. Salient input variables are specific for each fuzzy rule and are found during the genetic search process. It is shown through computer simulations of four real world problems that a large number of rules and input variables can be eliminated from the model without deteriorating the classification accuracy. By contrast, the classification accuracy of unseen data is increased due to the elimination.

show abstract

“…Ishibuchi and Nojimaa [38] examined the interpretability-accuracy tradeoff in fuzzy rule-based classifiers using a multiobjective fuzzy genetics-based machine learning (GBML) algorithm which is a hybrid version of Michigan and Pittsburgh approaches. Each fuzzy rule is represented by its antecedent fuzzy sets as an integer string of fixed length.…”

Section: Multiobjective Evolutionary Design Of Intelligent Paradigmsmentioning

confidence: 99%

Engineering Evolutionary Intelligent Systems: Methodologies, Architectures and Reviews

Abraham

Groşan

2008

Engineering Evolutionary Intelligent Systems

View full text Add to dashboard Cite

Summary. Designing intelligent paradigms using evolutionary algorithms is getting popular due to their capabilities in handling several real world problems involving complexity, noisy environment, imprecision, uncertainty and vagueness. In this Chapter, we illustrate the various possibilities for designing intelligent systems using evolutionary algorithms and also present some of the generic evolutionary design architectures that has evolved during the last couple of decades. We also provide a review of some of the recent interesting evolutionary intelligent system frameworks reported in the literature.

show abstract

Analysis of interpretability-accuracy tradeoff of fuzzy systems by multiobjective fuzzy genetics-based machine learning

Cited by 344 publications

References 25 publications

Fuzzy pattern trees as an alternative to rule-based fuzzy systems: Knowledge-driven, data-driven and hybrid modeling of color yield in polyester dyeing

Fuzzy pattern trees as an alternative to rule-based fuzzy systems: Knowledge-driven, data-driven and hybrid modeling of color yield in polyester dyeing

A general framework for designing a fuzzy rule-based classifier

Engineering Evolutionary Intelligent Systems: Methodologies, Architectures and Reviews

Contact Info

Product

Resources

About