Evolutionary adapted ensemble for reoccurring context

Authors:
Konrad Jackowski
Affiliations:
Departments of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland
Venue:
HAIS'12 Proceedings of the 7th international conference on Hybrid Artificial Intelligent Systems - Volume Part II
Year:
2012

Citing 7
Cited 0

The weighted majority algorithm

Information and Computation
Learning in the presence of concept drift and hidden contexts

Machine Learning
The impact of changing populations on classifier performance

KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Pasting Small Votes for Classification in Large Databases and On-Line

Machine Learning
Online Ensemble Learning

Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence
Mining concept-drifting data streams using ensemble classifiers

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Dynamic integration of classifiers for handling concept drift

Information Fusion

Quantified Score

Hi-index	0.00

Visualization

Abstract

Reoccurring Context is a phenomenon being subject of interest in machine learning theory dealing with Concept Drift. Periodic reappearance of contexts naturally encourage designing classifier systems which utilizes their expertize on contexts collected in the past. The paper presents study on EAERC algorithm that gather its knowledge on appearing contexts in form of elementary classifiers which can potentially contribute in ensemble classifier system if necessary while keeping ensemble size strictly limited to ensure short response time. While unseen context appears EAERC automatically adds new classifier to the pool.