A new algorithm for identifying fuzzy measures and its application to pattern recognition

@article{Grabisch1995ANA,
  title={A new algorithm for identifying fuzzy measures and its application to pattern recognition},
  author={Michel Grabisch},
  journal={Proceedings of 1995 IEEE International Conference on Fuzzy Systems.},
  year={1995},
  volume={1},
  pages={145-150 vol.1}
}
  • M. Grabisch
  • Published 20 March 1995
  • Computer Science
  • Proceedings of 1995 IEEE International Conference on Fuzzy Systems.
We present a new algorithm for identifying fuzzy measures, which is a kind of gradient algorithm with constraints. Its performance is superior to the one of previous attempts, and we show its efficiency to a problem of pattern recognition using Choquet integral.<<ETX>> 

Figures and Tables from this paper

New Pattern Recognition Tools Based on Fuzzy Logic for Image Understanding
TLDR
It is shown that fuzzy integrals and fuzzy rules provide powerful tools for solving pattern recognition problems in image understanding and object recognition in particular.
An Algorithm Based on Alternative Projections for a Fuzzy Measure Identification Problem
TLDR
The properties of the solution set of a fuzzy measure identification problem have been studied in detail and the efforts to develop an efficient algorithm to identify a fuzzy measures from examples have been made.
Fuzzy Integral for Classification and Feature Extraction
TLDR
The use of fuzzy integral is described, which can be viewed as a information fusion model, embedded into the framework of fuzzy pattern matching, and the problem of feature extraction is addressed.
Constructing fuzzy measures: a new method and its application to cluster analysis
  • Bo Yuan, G. Klir
  • Computer Science
    Proceedings of North American Fuzzy Information Processing
  • 1996
We first prove that for a given set of data there exists a fuzzy measure fitting exactly the data if and only if there exists an exact solution of the associated fuzzy relation equation. Secondly, we
A Fast Iterative Algorithm for Identifying Feature Scales and Signed Fuzzy Measures in Generalized Choquet Integrals
  • Xutao Deng, Zhenyuan Wang
  • Computer Science
    The 14th IEEE International Conference on Fuzzy Systems, 2005. FUZZ '05.
  • 2005
TLDR
An iterative algorithm for identifying approximate optimal scales for feature attributes of generalized Choquet integrals is developed and the optimal values of fuzzy measures are obtained immediately using least square method.
An algebraic method and a genetic algorithm to the identification of fuzzy measures based on Choquet integrals
TLDR
An algebraic method and a genetic algorithm to identify fuzzy measures based on the Choquet integral are introduced and some experimental results are presented on both artificial and real-world data sets to show their effectiveness.
Revised HLMS: A useful algorithm for fuzzy measure identification
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 13 REFERENCES
Classification by fuzzy integral: performance and tests
Multi-attribute classification using fuzzy integral
  • M. Grabisch, M. Sugeno
  • Computer Science
    [1992 Proceedings] IEEE International Conference on Fuzzy Systems
  • 1992
TLDR
A new method based on a fuzzy integral and possibility theory is presented, and a critical examination of the statistical approach and the supervised learning process is outlined.
Fuzzy integral in multicriteria decision making
Fuzzy Measure Theory
Introduction. Required Background in Set Theory. Fuzzy Measures. Extensions. Structural Characteristics for Set Functions. Measurable Functions on Fuzzy Measure Spaces. Fuzzy Integrals. PanIntegrals.
An Empirical Comparison of Pattern Recognition, Neural Nets, and Machine Learning Classification Methods
TLDR
For these problems, which have relatively few hypotheses and features, the machine learning procedures for rule induction or tree induction clearly performed best.
Fundamentals of Uncertainty Calculi with Applications to Fuzzy Inference
TLDR
The aim of this book is to provide a rigorous background for uncertainty calculi, with an emphasis on fuzziness, and to give a complete background on uncertainty measures and integrals, especially in a fuzzy setting.
Non-additive measure and integral
Preface. 1. Integration of Monotone Functions on Intervals. 2. Set Functions and Caratheodory Measurability. 3. Construction of Measures using Topology. 4. Distribution Functions, Measurability and
...
1
2
...