System identification (2nd ed.): theory for the user
System identification (2nd ed.): theory for the user
Medical Decision Support Systems
ISMDA '00 Proceedings of the First International Symposium on Medical Data Analysis
Fault Diagnosis: Models, Artificial Intelligence, Applications
Fault Diagnosis: Models, Artificial Intelligence, Applications
Feature Selection for Unsupervised Learning
The Journal of Machine Learning Research
Handling of inconsistent rules with an extended model of fuzzy reasoning
Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Evolving fuzzy classifiers using different model architectures
Fuzzy Sets and Systems
Evolving Intelligent Systems: Methodology and Applications
Evolving Intelligent Systems: Methodology and Applications
On-line incremental feature weighting in evolving fuzzy classifiers
Fuzzy Sets and Systems
Evolving Fuzzy Systems - Methodologies, Advanced Concepts and Applications
Evolving Fuzzy Systems - Methodologies, Advanced Concepts and Applications
Interpretability of linguistic fuzzy rule-based systems: An overview of interpretability measures
Information Sciences: an International Journal
Learning in Non-Stationary Environments: Methods and Applications
Learning in Non-Stationary Environments: Methods and Applications
An approach to online identification of Takagi-Sugeno fuzzy models
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Improving the interpretability of TSK fuzzy models by combining global learning and local learning
IEEE Transactions on Fuzzy Systems
A highly interpretable form of Sugeno inference systems
IEEE Transactions on Fuzzy Systems
An Evolving Fuzzy Predictor for Industrial Applications
IEEE Transactions on Fuzzy Systems
| Hi-index | 0.00 |
In this position paper, we are investigating interpretability issues in the context of evolving fuzzy systems (EFS). Current EFS approaches, developed during the last years, are basically providing methodologies for precise modeling tasks, i.e. relations and system dependencies implicitly contained in on-line data streams are modeled as accurately as possible. This is achieved by permanent dynamic updates and evolution of structural components. Little attention has been paid to the interpretable power of these evolved systems, which, however, originally was one fundamental strength of fuzzy models over other (data-driven) model architectures. This paper will present the (little) achievements already made in this direction, discuss new concepts and point out open issues for future research. Various well-known and important interpretability criteria will serve as basis for our investigations.