Jean-Luc Voz

Learn More
Many neural-like algorithms cummtly under study support classification tasks. Several of these algorithms base their functionality on LVQ-like procedures to find locations of centroids in the data space, and on kernel (or radial-basis) functions centered on these centroids to approximate functions or probability densities. A generic analog chip could(More)
For pattern classiication in a multi-dimensional space, the minimum misclassiication rate is obtained by using the Bayes criterion. Kernel estimators or probabilistic neural networks provide a good way t o e v aluate the probability densities of each class of data and are an interesting parallel implementation of the Bayesian classiier 1]. However, their(More)
Task B1 of the Elena project is aimed to provide a set of databases to be used for tests and benchmarks of the neural and classiication algorithms studied and developed in the project. This report describes all databases selected for this purpose. They are splitted into two parts: artiicially generated databases, mainly used for preliminary tests, and real(More)
  • 1