New results on observations selection

Authors:
R. T. Peres;C. E. Pedreira
Affiliations:
Electrical Engineering Department - PUC-Rio.;Faculty of Medicine and the Engineering Graduate Program (COPPE-PEE), Federal University of Rio de Janeiro, Brazil
Venue:
SMO'07 Proceedings of the 7th WSEAS International Conference on Simulation, Modelling and Optimization
Year:
2007

Citing 4
Cited 0

Vector quantization and signal compression

Vector quantization and signal compression
Generalization error estimates and training data valuation

Generalization error estimates and training data valuation
Learning Vector Quantization with Training Data Selection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Improving generalization by data categorization

PKDD'05 Proceedings of the 9th European conference on Principles and Practice of Knowledge Discovery in Databases

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this paper we propose a new observations selection technique in Vector Quantization context. The main idea is to select observations that are not representatives of their classes. The cells generated by the Vector Quantization are divided in regions of rejection. A number of controlled experiments were performed demonstrating the proposed methodology potentiality.