Granular instances selection for fuzzy modeling

Authors:
S. Sakinah S. Ahmad;Witold Pedrycz
Affiliations:
Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada;Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada
Venue:
MLDM'11 Proceedings of the 7th international conference on Machine learning and data mining in pattern recognition
Year:
2011

Citing 10
Cited 0

The KDD process for extracting useful knowledge from volumes of data

Communications of the ACM
Reduction Techniques for Instance-BasedLearning Algorithms

Machine Learning
On Issues of Instance Selection

Data Mining and Knowledge Discovery
Fuzzy Systems Engineering: Toward Human-Centric Computing

Fuzzy Systems Engineering: Toward Human-Centric Computing
On the use of the weighted fuzzy c-means in fuzzy modeling

Advances in Engineering Software
Data-driven fuzzy modeling for Takagi-Sugeno-Kang fuzzy system

Information Sciences: an International Journal
IFS-CoCo: Instance and feature selection based on cooperative coevolution with nearest neighbor rule

Pattern Recognition
A review of instance selection methods

Artificial Intelligence Review
Using evolutionary algorithms as instance selection for data reduction in KDD: an experimental study

IEEE Transactions on Evolutionary Computation
MembershipMap: Data Transformation Based on Granulation and Fuzzy Membership Aggregation

IEEE Transactions on Fuzzy Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Instances selection is an important task in the data preparation phase of knowledge discovery and data mining (KDD). Instances selection techniques are largely studied especially in the classification problem. However, little work has been done to implement instances selection in fuzzy modeling application. In this paper, we present a framework for fuzzy modeling using the granular instances selection. This method is based on the information granulation approach to select the best subset of instances for constructing the fuzzy model. We show that by using Particle Swarm Optimization (PSO) for searching the best level of granularity for each feature can improve the predictive accuracy of the fuzzy model.