In this paper, we propose to use bootstrap samples in designing an extended quadratic classifier. The proposed method is to generate bootstrap samples, which have much information about distributions, and to optimize the extended quadratic classifier so that the error estimated by using bootstrap samples is minimized. The performance of the proposed classifier is demonstrated on real data.
Citation:
T. Miyamoto, Y. Hamamoto, Y. Mitani, "Use of Bootstrap Samples in Quadratic Classifier Design," icpr, vol. 2, pp.2789, 15th International Conference on Pattern Recognition (ICPR'00) - Volume 2, 2000