Currently displaying 1 – 3 of 3

Showing per page

Order by Relevance | Title | Year of publication

Piecewise linear classifiers preserving high local recognition rates

Hiroshi Tenmoto;  Mineichi Kudo;  Masaru Shimbo — 1998

Kybernetika

We propose a new method to construct piecewise linear classifiers. This method constructs hyperplanes of a piecewise linear classifier so as to keep the correct recognition rate over a threshold for a training set. The threshold is determined automatically by the MDL (Minimum Description Length) criterion so as to avoid overfitting of the classifier to the training set. The proposed method showed better results in some experiments than a previous method.

Construction of nonlinear discrimination function based on the MDL criterion

Manabu Sato;  Mineichi Kudo;  Jun Toyama;  Masaru Shimbo — 1998

Kybernetika

Although a nonlinear discrimination function may be superior to linear or quadratic classifiers, it is difficult to construct such a function. In this paper, we propose a method to construct a nonlinear discrimination function using Legendre polynomials. The selection of an optimal set of Legendre polynomials is determined by the MDL (Minimum Description Length) criterion. Results using many real data show the effectiveness of this method.

Page 1

Download Results (CSV)