Summary
International Symposium on Nonlinear Theory and its Applications
2005
Session Number:2-1-5
Session:
Number:2-1-5-3
Cellular analysis of covariance structure for Data Mining by Backward Euler method
Yuko Zennyoji, Nao Ohashi, Masayuki Yamauchi, Mamoru Tanaka,
pp.341-344
Publication Date:2005/10/18
Online ISSN:2188-5079
DOI:10.34385/proc.40.2-1-5-3
PDF download (125.7KB)
Summary:
This paper describes a cellular analysis of covariance structure for data mining using Back Euler method. It is the implicit method which is the most practical method for solving stiff systems. It is difficult to solve these systems with conventional methods. Davidon-Fletcher-Powell (DFP) method, one of quasi-Newton methods, is utilized to modify the next step solution. At each iteration step for quasi-Newton method, the approximation to the matrix including second-order partial derivatives is updated by using new gradient information. By using both of Back Euler method and DFP method, the solution for stiff systems in the parameter space can be obtained.