A search engine based on neural correlation matrix memories

This paper describes a novel search and pattern matching technology, AURA, based on neural networks. The technology exploits the simple and fast train ing of correlation matrix memories and their ability to match noisy and inc omplete data. Unlike other neural network approaches, the method scales wel l to accommodate very large datasets, and ha's a simple and effective hardw are implementation. It achieves this by storing basic feature data and rely ing on the fast matching ability of correlation matrix memory methods, as w ell as a two-stage processing approach. The paper demonstrates that the tec hnology is applicable to a wide range of problems and shows how it has been implemented in dedicated hardware for high-performance applications. (C) 2 000 Elsevier Science B.V. All rights reserved.