Learn More
This paper studies Regularized Discriminant Analysis (RDA) in the context of face recognition. We check RDA sensitivity to different photometric preprocessing methods and compare its performance to other classifiers. Our study shows that RDA is better able to extract the relevant discriminatory information from training data than the other classifiers(More)
– In this paper we seek a Gaussian mixture model (GMM) of the class-conditional densities for plug-in Bayes classification. We propose a method for setting the number of the components and the covariance matrices of the class-conditional GMMs. It compromises between simplicity of the model selection based on the Bayesian information criterion (BIC) and the(More)
Midazolam (Versed), the first water-soluble benzodiazepine, has had widespread acceptance as a parenteral anxiolitic agent. Its antiepileptic properties were studied in adult patients with good results. Midazolam was administered intramuscularly to 48 children, ages 4 months to 14 years, with 69 epileptic episodes of various types. In all but 5 epileptic(More)
Sammon's mapping is conventionally used for exploratory data projection, and as such is usually inapplicable for classification. In this paper we apply a neural network (NN) implementation of Sammon's mapping to classification by extracting an arbitrary number of projections. The projection map and classification accuracy of the mapping are compared with(More)
The human beta-globin locus control region (LCR) controls the transcription, chromatin structure, and replication timing of the entire locus. DNA replication was found to initiate in a transcription-independent manner within a region located 50 kilobases downstream of the LCR in human, mouse, and chicken cells containing the entire human beta-globin locus.(More)
The projection maps and derived classification accuracies of a neural network (NN) implementation of Sammon's mapping, an auto-associative NN (AANN) and a multilayer perceptron (MLP) feature extractor are compared with those of the conventional principal component analysis (PCA). Tested on five real-world databases, the MLP provides the highest(More)
Brainstem auditory evoked potentials (BAEPs) were performed on 114 children with attention deficit disorder (ADD). Prolonged latencies of waves III and V and longer brainstem transmission time interval of waves I-III and I-V were observed in the study group compared with normal controls. A significant asymmetry of wave III latency between the ears was found(More)
List of figures Fig.1. Normal class-conditional distributions. Test error rates in the ExF plots (β 1 =0). Bars represent the standard errors of the error rates. Stars (***) indicate the existence of significant differences between the averaged errors of REM and ORTH (t-test, p=0.05). Fig.2. Nonnormal unimodal class-conditional distributions. Test error(More)
A recently published generalized growing and pruning (GGAP) training algorithm for radial basis function (RBF) neural networks is studied and modified. GGAP is a resource-allocating network (RAN) algorithm, which means that a created network unit that consistently makes little contribution to the network's performance can be removed during the training.(More)