Home  About us  Contact
 

Multilayer Perceptron Neural Networks (multilayer + perceptron_neural_network)

Distribution by Scientific Domains
Chemistry75%

Selected Abstracts

A comparative study of multilayer perceptron neural networks for the identification of rhubarb samples

PHYTOCHEMICAL ANALYSIS, Issue 2 2007
Zhuoyong Zhang
Abstract Artificial neural networks have gained much attention in recent years as fast and flexible methods for quality control in traditional medicine. Near-infrared (NIR) spectroscopy has become an accepted method for the qualitative and quantitative analyses of traditional Chinese medicine since it is simple, rapid, and non-destructive. The present paper describes a method by which to discriminate official and unofficial rhubarb samples using three layer perceptron neural networks applied to NIR data. Multilayer perceptron neural networks were trained with back propagation, delta-bar-delta and quick propagation algorithms. Results obtained using these methods were all satisfactory, but the best outcomes were obtained with the delta-bar-delta algorithm. Copyright © 2006 John Wiley & Sons, Ltd. [source]

Strategies for preventing defection based on the mean time to defection and their implementations on a self-organizing map

EXPERT SYSTEMS, Issue 5 2005
Young Ae Kim
Abstract: Customer retention is a critical issue for the survival of any business in today's competitive marketplace. In this paper, we propose a dynamic procedure utilizing self-organizing maps and a Markov process for detecting and preventing customer defection that uses data of past and current customer behavior. The basic concept originates from empirical observations that identified that a customer has a tendency to change behavior (i.e. trim-out usage volumes) before eventual withdrawal and defection. Our explanatory model predicts when potential defectors are likely to withdraw. Two strategies are suggested to respond to the question of where to lead potential defectors for the next stage, based on anticipating when the potential defector will leave. Our model predicts potential defectors with little deterioration of prediction accuracy compared with that of the multilayer perceptron neural network and decision trees. Moreover, it performs reasonably well in a controlled experiment using an online game. [source]

A comparative study of multilayer perceptron neural networks for the identification of rhubarb samples

PHYTOCHEMICAL ANALYSIS, Issue 2 2007
Zhuoyong Zhang
Abstract Artificial neural networks have gained much attention in recent years as fast and flexible methods for quality control in traditional medicine. Near-infrared (NIR) spectroscopy has become an accepted method for the qualitative and quantitative analyses of traditional Chinese medicine since it is simple, rapid, and non-destructive. The present paper describes a method by which to discriminate official and unofficial rhubarb samples using three layer perceptron neural networks applied to NIR data. Multilayer perceptron neural networks were trained with back propagation, delta-bar-delta and quick propagation algorithms. Results obtained using these methods were all satisfactory, but the best outcomes were obtained with the delta-bar-delta algorithm. Copyright © 2006 John Wiley & Sons, Ltd. [source]

Estimation of vapour,liquid equilibrium data for binary refrigerant systems containing 1,1,1,2,3,3,3-heptafluoropropane (R227ea) by using artificial neural networks

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, Issue 2 2010
M. R. Nikkholgh
Abstract In this research, the ability of multilayer perceptron neural networks to estimate vapour,liquid equilibrium data have been studied. Four typical binary refrigerant systems containing R227ea have been investigated in a large range of temperatures and pressures. The systems are categorised into four groups, based on their different deviations from the Raoult's law. The networks with one hidden layer consisted of five neurons are developed as the optimal structure. For these binary systems, uncertainties in the artificial neural networks (ANNs) estimations were not more than 1.03%. In addition, the abilities of ANNs are shown by comparisons with Margules, van Laar, and some other correlations. Dans ce travail de recherche, nous avons étudié la capacité de réseaux neuraux de perceptron multicouche à estimer les données d'équilibre vapeur-liquide. Quatre systèmes typiques de réfrigérants binaires contenant du R227ea ont été étudiés sur de grands intervalles de température et de pression. Les systèmes étaient classés en quatre groupes, en fonction de leurs déviations différentes par rapport à la Loi de Raoult. Les réseaux ayant une couche cachée composée de cinq neurones sont développés comme la structure optimale. Pour ces systèmes binaires, les incertitudes dans les estimations ANN ne dépassaient pas 1,029 %. De plus, les capacités des ANN sont données en comparaison avec Margules, van Laar et certaines autres corrélations. [source]