A comparative analysis between conventional approaches and connectionist methods in pattern recognition tasks
Oswaldo Ludwig, Leizer Schnitman, Jamf de Souza, Herman Augusto Lepikson · 2004
The purpose of this work is to compare statistical and connectionist techniques for patterns recognition. Connectionist approach is based on feedforward artificial neural network, self-organizing maps or hybrid algorithms which are also compared. An alternative preprocessing method to pattern recognition problems is also suggested.