A simple additive non-parametric regression method and its applic | 667
Drug Designing: Open Access

Drug Designing: Open Access
Open Access

ISSN: 2169-0138

+44 1223 790975

A simple additive non-parametric regression method and its application to QSAR

International Conference and Exhibition on Computer Aided Drug Design & QSAR

October 29-31, 2012 DoubleTree by Hilton Chicago-North Shore, USA

Tatsuya Takagi, Akiko Miyamoto, Rika Nishikiori, Masaya Kawase, Kousuke Okamoto, Norihito Kawashita, Asuka Hatabu, Haibo Wang and Teruo Yasunaga

Scientific Tracks Abstracts: Drug Design

Abstract :

Various kinds of smoothing methods as well as non-parametric regression methods have been proposed and applied to various kinds of fields. For example MARS (Multivariate Adaptive Regression Splines), which were proposed by J. H. Friedman, has been widely used. Lowess (locally weighted scatterplot smoothing) proposed by W. S. Cleveland has been also widely utilized. However, most of them have some imperfections. For instance, Lowess is not easy to extend the case of using multiple predictor variables. In addition, MARS like methods are so complex that other parameters such as AUX cannot be obtained easily because their main purposes are smoothing.

Biography :

Tatsuya Takagi graduated School of Pharmaceutical Sciences, Osaka University in 1979 and became a research assistant at the same department in 1980. He has completed his Ph.D at the age of 32 years from Osaka University and became a lecturer at Genoe Information Research Center, Osaka University in 1993. Since 1998, he has been the Professor of Lab. of Environmental Pharmacometrics, Graduate School of Pharmaceutical Sciences, Osaka University. He has published more than 100 papers in reputed journals and serving as an editorial board member of repute.