Adaptive Mechanisms for Classification Problems with Drifting Data

Authors:
Zoheir Sahel;Abdelhamid Bouchachia;Bogdan Gabrys;Paul Rogers
Affiliations:
School of Design, Engineering and Computing, Bournemouth University, UK;Dept. of Informatics, University of Klagenfurt, Austria;School of Design, Engineering and Computing, Bournemouth University, UK;School of Design, Engineering and Computing, Bournemouth University, UK
Venue:
KES '07 Knowledge-Based Intelligent Information and Engineering Systems and the XVII Italian Workshop on Neural Networks on Proceedings of the 11th International Conference
Year:
2007

Citing 8
Cited 3

A Nearest Hyperrectangle Learning Method

Machine Learning
Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system

Neural Networks
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
Towards Incremental Fuzzy Classifiers

Soft Computing - A Fusion of Foundations, Methodologies and Applications
Incremental Learning By Decomposition

ICMLA '06 Proceedings of the 5th International Conference on Machine Learning and Applications
Online pattern classification with multiple neural network systems: an experimental study

IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
General fuzzy min-max neural network for clustering and classification

IEEE Transactions on Neural Networks

Incremental learning with multi-level adaptation

Neurocomputing
Distributed learning with data reduction

Transactions on computational collective intelligence IV
Application of agent-based simulated annealing and tabu search procedures to solving the data reduction problem

International Journal of Applied Mathematics and Computer Science - Semantic Knowledge Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Most work on supervised learning is undertaken on static problems. However, in many real world classification problems, the environment in which the classifiers operate is dynamic i.e. the descriptions of classes change with time. In this paper, the process of generating drifting data is introduced in order to assess two adaptive approaches that deal with dynamically changing data. These approaches are: retraining on evolving data set and incremental learning. The empirical evaluation has shown that both these approaches improve the performance compared to the non-adaptive mode though a number of outstanding research issues remain.