Limited Sample Based Optimum Classifier Design and the Evaluation of the Mean Error Rate
スポンサーリンク
概要
- 論文の詳細を見る
This paper deals with limited sample based optimum classifier design and the theoretical evaluation of the mean error rate. Gaussian population with unknown parameters is assumed. The conditional density given a limited sample of the population is first derived, and its relationship to the multivariate t-distribution is shown. Then, the mean error rate of the optimum classifier is theoretically evaluated by the quadrature of the conditional density. To verify the optimality of the classifier and correctness of the mean error calculation, the results of Monte Carlo simulation employing a new sampling procedure are shown. The role of the a priori distribution in reducing the mean error rate is discussed at the end of this paper.
- Faculty of Engineering, Mie Universityの論文
- 1998-12-25
Faculty of Engineering, Mie University | 論文
- Approximate Method of Elastic Buckling Strength Analysis for Irregular Frames
- Elastic Buckling Strength of Steel Frames Connected to Core Wall
- Calculations of Energy Shift of the Conduction Band-Edge in Doped and Compensated GaP
- Estimation of the Glass Transition Pressure of Liquids under Very High Pressure by Viscoelastic Analysis
- Information Extraction from Electronic Mail