期刊
NEUROCOMPUTING
卷 55, 期 1-2, 页码 337-346出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/S0925-2312(03)00374-6
关键词
SVM; support vector machines; simplex optimisation; cross-validation; human intestinal absorption; blood-brain barrier; QSAR; regression; classification; variable selection
In this work, the performance and predictive capability of support vector machine (SVM) algorithms are investigated within the area of pharmaceutical drug design. The investigations in this paper clearly indicate the crucial importance of SVM parameter optimisation as well as variable selection in order to develop statistical models with good predictive capabilities on external test sets when using SVM regression. The simplex optimisation technique is applied to both traditional quantitative structure-activity relationships as well as important drug transport processes such as intestinal absorption and blood-brain barrier partitioning. (C) 2003 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据