Recommending internet-domains using trails and neural networks

Abstract

This paper discusses the use of artificial neural networks, trained with patterns extracted from trail data, as recommender systems. Feed-forward Multilayer-Perceptrons trained with the Backpropagation Algorithm were used to assign a rating to pairs of domains, based on the number of people that had traversed between them. The artificial neural network constructed in this project was capable of learning the training set to a great extent, and showed good generalizational capacities. © 2002 Springer-Verlag Berlin Heidelberg.