Title :
Using support vector machines for mining regression classes in large data sets
Author :
Sun, Zonghai ; Gao, Lixin ; Sun, Youxian
Author_Institution :
Nat. Lab. of Ind. Control Technol., Zhejiang Univ., Hangzhou, China
Abstract :
Support vector machines (SVM) overcome the limit of the maximum-likelihood. method that only applies to very limited set of the density functions. They can estimate simultaneously the regression classes in the mixture data set. The validity of the SVM was demonstrated in experiments. The results indicate that the SVM can estimate the regression classes in the mixture data set with noise.
Keywords :
data mining; regression analysis; support vector machines; very large databases; SVM; data mining; large data sets; mixture data set; noise; regression classes; support vector machines; Data mining; Density functional theory; Educational institutions; Industrial control; Kernel; Laboratories; Pattern classification; Polynomials; Support vector machine classification; Support vector machines;
Conference_Titel :
TENCON '02. Proceedings. 2002 IEEE Region 10 Conference on Computers, Communications, Control and Power Engineering
Print_ISBN :
0-7803-7490-8
DOI :
10.1109/TENCON.2002.1181221
