A search engine based on neural correlation matrix memories

Jim Austin; Ken Lees

A search engine based on neural correlation matrix memories

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

This paper describes a novel search and pattern matching technology, AURA, based on neural networks. The technology exploits the simple and fast training of correlation matrix memories and their ability to match noisy and incomplete data. Unlike other neural network approaches, the method scales well to accommodate very large datasets, and ha's a simple and effective hardware implementation. It achieves this by storing basic feature data and relying on the fast matching ability of correlation matrix memory methods, as well as a two-stage processing approach. The paper demonstrates that the technology is applicable to a wide range of problems and shows how it has been implemented in dedicated hardware for high-performance applications. (C) 2000 Elsevier Science B.V. All rights reserved.

Original language	English
Pages (from-to)	55-72
Number of pages	18
Journal	Neurocomputing
Volume	35
Issue number	1-4
Publication status	Published - Nov 2000

Keywords

AURA
correlation matrix memory
neural networks
pattern recognition
search engines
ASSOCIATIVE MEMORY

Access to Document

http://dx.doi.org/10.1016/S0925-2312(00)00309-X

Cite this

@article{e3a5a98746854edda0d8097db1946e29,

title = "A search engine based on neural correlation matrix memories",

abstract = "This paper describes a novel search and pattern matching technology, AURA, based on neural networks. The technology exploits the simple and fast training of correlation matrix memories and their ability to match noisy and incomplete data. Unlike other neural network approaches, the method scales well to accommodate very large datasets, and ha's a simple and effective hardware implementation. It achieves this by storing basic feature data and relying on the fast matching ability of correlation matrix memory methods, as well as a two-stage processing approach. The paper demonstrates that the technology is applicable to a wide range of problems and shows how it has been implemented in dedicated hardware for high-performance applications. (C) 2000 Elsevier Science B.V. All rights reserved.",

keywords = "AURA, correlation matrix memory, neural networks, pattern recognition, search engines, ASSOCIATIVE MEMORY",

author = "Jim Austin and Ken Lees",

year = "2000",

month = nov,

language = "English",

volume = "35",

pages = "55--72",

journal = "Neurocomputing",

publisher = "Elsevier",

number = "1-4",

}

TY - JOUR

T1 - A search engine based on neural correlation matrix memories

AU - Austin, Jim

AU - Lees, Ken

PY - 2000/11

Y1 - 2000/11

N2 - This paper describes a novel search and pattern matching technology, AURA, based on neural networks. The technology exploits the simple and fast training of correlation matrix memories and their ability to match noisy and incomplete data. Unlike other neural network approaches, the method scales well to accommodate very large datasets, and ha's a simple and effective hardware implementation. It achieves this by storing basic feature data and relying on the fast matching ability of correlation matrix memory methods, as well as a two-stage processing approach. The paper demonstrates that the technology is applicable to a wide range of problems and shows how it has been implemented in dedicated hardware for high-performance applications. (C) 2000 Elsevier Science B.V. All rights reserved.

AB - This paper describes a novel search and pattern matching technology, AURA, based on neural networks. The technology exploits the simple and fast training of correlation matrix memories and their ability to match noisy and incomplete data. Unlike other neural network approaches, the method scales well to accommodate very large datasets, and ha's a simple and effective hardware implementation. It achieves this by storing basic feature data and relying on the fast matching ability of correlation matrix memory methods, as well as a two-stage processing approach. The paper demonstrates that the technology is applicable to a wide range of problems and shows how it has been implemented in dedicated hardware for high-performance applications. (C) 2000 Elsevier Science B.V. All rights reserved.

KW - AURA

KW - correlation matrix memory

KW - neural networks

KW - pattern recognition

KW - search engines

KW - ASSOCIATIVE MEMORY

M3 - Article

VL - 35

SP - 55

EP - 72

JO - Neurocomputing

JF - Neurocomputing

IS - 1-4

ER -