Documents, concepts and neural networks
Jennifer Farkas · 1993
In this paper we investigate the relevance of neural networks to the problem of document classification. We show that textual documents can be represented numerically in a semantically meaningful way, so that the back-propagation learning algorithm can be used to build a document classifying neural network. We show that the network can be taught to classify natural language text according to predefined specifications. The convergence properties of the prototype NeuroZ described in this paper make it clear that neural networks provide a new platform for the automatic classification of semantically similar documents and that a system can be built which distinguishes between relatively complex linguistic patterns.