Evolving fuzzy classifier based on the modified ECM algorithm for pattern classification

Authors:
Maurílio J. Inácio;Renato D. Maia;Walmir M. Caminhas
Affiliations:
Computer Engineering Dept., FACIT, Montes Claros, MG, Brazil,Computer Sciences Dept., UNIMONTES, Montes Claros, MG, Brazil;Automation and Control Engineering Dept., FACIT, Montes Claros, MG, Brazil,Computer Sciences Dept., UNIMONTES, Montes Claros, MG, Brazil,Institute of Agricultural Sciences, UFMG, Montes Claros, MG ...;Electronic Engineering Dept., UFMG, Belo Horizonte, MG, Brazil
Venue:
IDEAL'12 Proceedings of the 13th international conference on Intelligent Data Engineering and Automated Learning
Year:
2012

Citing 8
Cited 0

Evolving rule-based models: a tool for design of flexible adaptive systems

Evolving rule-based models: a tool for design of flexible adaptive systems
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Evolving fuzzy classifiers using different model architectures

Fuzzy Sets and Systems
Data clustering: 50 years beyond K-means

Pattern Recognition Letters
Evolving Intelligent Systems: Methodology and Applications

Evolving Intelligent Systems: Methodology and Applications
Evolving Fuzzy Systems - Methodologies, Advanced Concepts and Applications

Evolving Fuzzy Systems - Methodologies, Advanced Concepts and Applications
DENFIS: dynamic evolving neural-fuzzy inference system and its application for time-series prediction

IEEE Transactions on Fuzzy Systems
FLEXFIS: A Robust Incremental Learning Approach for Evolving Takagi–Sugeno Fuzzy Models

IEEE Transactions on Fuzzy Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Nowadays, online and real-time pattern classification applications are required in many areas. Most classification algorithms are suitable only for off-line applications. Using the concept of evolving intelligent systems, this paper proposes an evolving fuzzy classifier capable of creating the rule base in online mode and real-time. The proposed evolving fuzzy classifier is based on a new clustering algorithm that consists of an improved version of the Evolving Clustering Method (ECM). Experiments with well-known benchmark classification problems indicated that the proposal is promising.