Full Text
(53 K)

Prediction of Rotor Spun Cotton Yarn Quality: A Comparison of Neural Network and Regression Algorithms

Dean Ethridge, Reiyao Zhu


Neural Networks provide an alternative approach to understanding and predicting complex relationships among fiber properties and their impacts on spinning performance and textile product quality. This paper provides preliminary results on the relative performance of the "back-propagation" neural network algorithm versus linear regression. The application is to prediction of selected yarn properties based on instrument measurements of fifteen fiber properties. Results suggest that marginal improvements in predictive performance are possible with neural networks.

Reprinted from Proceedings of the 1996 Beltwide Cotton Conferences pp. 1314 - 1317
©National Cotton Council, Memphis TN

[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page
Document last modified Sunday, Dec 6 1998