Feature selection using multi-objective genetic algorithms for handwritten digit recognition

Feature selection using multi-objective genetic algorithms for handwritten digit recognition

Oliveira, L. S. and Sabourin, R. and Bortolozzi, F. and Suen, C. Y.

Proceedings – International Conference on Pattern Recognition 2002

Abstract : This paper discusses the use of genetic algorithm for feature selection for handwriting recognition. Its novelty lies in the use of a multi-objective genetic algorithms where sensitivity analysis and neural network are employed to allow the use of a representative database to evaluate fitness and the use of a validation database to identify the subsets of selected features that provide a good generalization. Comprehensive experiments on the NIST database confirm the effectiveness of the proposed strategy. © 2002 IEEE.