Prediction Accuracy of Neural Network Models

  • G. Rutka Riga Technical University

Abstract

A view of models used for Internet traffic prediction using neural network applications is presented. We look at the problem of traffic prediction in the presence of self-similarity. Our experiments inspect prediction error of linear, multilayer perceptron and radial basis function networks. Ill. 5, bibl. 12 (in English; summaries in English, Russian and Lithuanian).

Published
2015-03-27
Section
T 180 TELECOMMUNICATION ENGINEERING