Inspired by a new sliding window-based adaptive fuzzy measure, this paper presents an application of the Adaptive Power Measure (APM) within Fuzzy Rule-Based Classification Systems (FRBCS) to improve the classification accuracy. The APM is used in the context of the Choquet integral and its generalizations to handle data aggregation in the fuzzy reasoning method of FRBCS. The APM dynamically adjusts the values of the power measure without relying on genetic algorithms (as usually done in the literature), using heuristics to define the exponent q of the standard Power Measure, which is adapted to class-specific characteristics. The methodology includes extensive testing on 33 datasets, employing various heuristic functions for determing q, and examining eight Choquet integral generalizations. Results reveal that the APM frequently matches or outperforms established methods, especially when the effects of the interaction among features are critical. Statistical analyses using the aligned Friedman rank test and Holm’s post-hoc test validate these improvements, establishing the APM as a robust and versatile approach for FRBCS applications.
On the application of the adaptive Power Measure in a Fuzzy Rule-Based Classification System
Giancarlo Lucca,G. Dimuro,H. Santos,T. Asmus,A. Yamin,R. H. S. Reiser,H. Camargo,C. Marco-Detchart,H. Bustince
Published 2025 in IEEE International Conference on Fuzzy Systems
ABSTRACT
PUBLICATION RECORD
- Publication year
2025
- Venue
IEEE International Conference on Fuzzy Systems
- Publication date
2025-07-06
- Fields of study
Mathematics, Computer Science
- Identifiers
- External record
- Source metadata
Semantic Scholar
CITATION MAP
EXTRACTION MAP
CLAIMS
- No claims are published for this paper.
CONCEPTS
- No concepts are published for this paper.
REFERENCES
Showing 1-36 of 36 references · Page 1 of 1
CITED BY
- No citing papers are available for this paper.
Showing 0-0 of 0 citing papers · Page 1 of 1