Importance of information pre‐processing in the improvement of neural network results*

: This paper compares the success ratio of certain topologies when their input data are changed through different pre-processing methods. It begins with the database description, and it shows some different kinds of pre-processing that will be applied and the necessary modifications to the input layer of the network. The process is carried out using four networks with supervised learning (Standard Backpropagation, Quick propagation, Resilient Propagation and Backpropagation with Momentum) and two with unsupervised learning (ART 1 and Dynamic Learning Vector Quantization).