Tetrahedron Computer Methodology - Vol. 3, Nos 3/4, 1990 - 5 db Floppy lemezzel [antikvár]

Tetrahedron Computer Methodology, Vol. 3, No. 3/4, pp. 119 to 128, 1990 Printed in Great Britain0898-5529/90 $3.00^.00 Pergamon Press picNeural Network Technology and its Application in Chemical ResearchMark E. LacyProduct Development, Norwich Ealon Pharmaceuticals, Inc., A Procter Gamble Company, P.O. Box 191, Norwich, New York 13815 USAReceived 11 September 1990, Revised 26 October, 1990, Accepted 26 October 1990Key words: Neural network; Computer methodology; Protein structure; Organic chemistry; Spectroscopy;Nucleic acidAbstract: Neural network technology is finding new applications in chemical research. This paper briefly describes neural network technology, reviews recent applications in chemistry, and introduces the papers appearing in this special issue of Tetrahedron Computer Methodology.INTRODUCTIONNeural network technology is the design and use of neural networks to solve computational problems. The roots of this technology were established in the 1940s and 1950s with the work of McCulloch Pitts1 on networks of idealized neuron-like elements, the development of the perception element by Frank Rosenblatt at Cornell, and the development of the "adaptive linear element" by Widrow.2 Funding for neural network research was sharply curtailed subsequent to Minsky Papert's analysis of the perceptron and description of its limitations.3 Several groups continued to work on improving neural networks and their capabilities, but it was not until the 1980s, when Hopfield4 published a clear description of a neural computing system whose interconnected elements seek an energy minimum, that interest in neural networks was revitalized. Suddenly neural network technology received widespread attention. The outstanding progress in computing power that had taken place since the 1960s, as well as the progress made in artificial intelligence on the expert systems front, each played a role in neural network technology becoming an established and sustained research focus. Now neural network technology is a flourishing field, with a number of journals {e.g., Neural Networks and International Journal of Neural Systems) and professional societies (e.g., the International Neural Network Society) devoted to it.The purpose of this paper is to (1) provide some information on the fundamentals of neural networks for those who are not familiar with the technology, (2) review neural network applications in chemistry that have been published over the last few years, and (3) briefly introduce the papers presented in this issue. Readers will find more extensive explanation of some neural network concepts in the articles in this issue, including