Introduction: The dataset of 826 patients who were suspected of the prostate cancer was examined. The best single marker and the combination of markers which could predict the prostate cancer in very early stage of the disease were looked for. Methods: For combination of markers the logistic regression, the multilayer perceptron neural network and the k-nearest neighbour method were used. 10 models for each method were developed on the training data set and the predictive accuracy verified on the test data set. Results and conclusions: The ROCs for the models were constructed and AUCs were estimated. All three examined methods have given comparable results. The medians of estimates of AUCs were 0.775, which were larger than AUC of the best single marker.
This work concentrates on a novel method for empirical estimation
of generalization ability of neural networks. Given a set of training (and testing) data, one can choose a network architecture (nurnber of layers, number of neurons in each layer etc.), an initialization method, and a learning algorithrn to obtain a network. One measure of the performance of a trained network is how dosely its actual output approximates the desired output for an input that it has never seen before. Current methods provide a “number” that indicates the estimation of the generalization ability of the network. However, this number provides no further inforrnation to understand the contributing factors when the generalization ability is not very good. The method proposed uses a number of parameters to define the generalization ability. A set of the values of these parameters provide an estimate of the generalization ability. In addition, the value of each pararneter indicates the contribution of such factors as network architecture, initialization method, training data set, etc. Furthermore, a method has been developed to verify the validity of the estimated values of the parameters.
The most frequently used instrument for measuring velocity distribution in the cross-section of small rivers is the propeller-type current meter. Output of measuring using this instrument is point data of a tiny bulk. Spatial interpolation of measured data should produce a dense velocity profile, which is not available from the measuring itself. This paper describes the preparation of interpolation models. Measuring campaign was realized to obtain operational data. It took place on real streams with different velocity distributions. Seven data sets were obtained from four cross-sections varying in the number of measuring points, 24-82. Following methods of interpolation of the data were used in the same context: methods of geometric interpolation arithmetic mean and inverse distance weighted, the method of fitting the trend to the data thin-plate spline and the geostatistical method of ordinary kriging. Calibration of interpolation models carried out in the computational program Scilab is presented. The models were tested with error criteria by cross-validation. Ordinary kriging was proposed to be the most suitable interpolation method, giving the lowest values of used error criteria among the rest of the interpolation methods.