Improving Classification under Changes in Class and Within-Class Distributions

Authors:
Rocío Alaiz-Rodríguez;Alicia Guerrero-Curieses;Jesús Cid-Sueiro
Affiliations:
Dpto. de Ingeniería Eléctrica y de Sistemas, Universidad de León, León, Spain 24071;Dpto. de Teoría de la Señal y Comunicaciones, Universidad Rey Juan Carlos., Fuenlabrada-Madrid, Spain 28943;Dpto. de Teoría de la Señal y Comunicaciones, Universidad Carlos III de Madrid., Leganés-Madrid, Spain
Venue:
IWANN '09 Proceedings of the 10th International Work-Conference on Artificial Neural Networks: Part I: Bio-Inspired Systems: Computational and Ambient Intelligence
Year:
2009

Citing 8
Cited 3

Growing Gaussian mixtures network for classification applications

Signal Processing
Robust Classification for Imprecise Environments

Machine Learning
Adjusting the outputs of a classifier to new a priori probabilities: a simple procedure

Neural Computation
On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions

Machine Learning
A Response to Webb and Ting's On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions

Machine Learning
Minimax Regret Classifier for Imprecise Class Distributions

The Journal of Machine Learning Research
Asymptotic Bayesian generalization error when training and test distributions are different

Proceedings of the 24th international conference on Machine learning
A model selection algorithm for a posteriori probability estimation with neural networks

IEEE Transactions on Neural Networks

A unifying view on dataset shift in classification

Pattern Recognition
Analysis of preprocessing vs. cost-sensitive learning for imbalanced classification. Open problems on intrinsic data characteristics

Expert Systems with Applications: An International Journal
Imprecise imputation as a tool for solving classification problems with mean values of unobserved features

Advances in Artificial Intelligence

Quantified Score

Hi-index	0.01

Visualization

Abstract

The fundamental assumption that training and operational data come from the same probability distribution, which is the basis of most learning algorithms, is often not satisfied in practice. Several algorithms have been proposed to cope with classification problems where the class priors may change after training, but they can show a poor performance when the class conditional data densities also change. In this paper, we propose a re-estimation algorithm that makes use of unlabeled operational data to adapt the classifier behavior to changing scenarios. We assume that (a) the classes may be decomposed in several (unknown) subclasses, and (b) the prior subclass probabilities may change after training. Experimental results with practical applications show an improvement over an adaptive method based on class priors, while preserving a similar performance when there are no subclass changes.