The Inverse Problem of Support Vector Machines Solved by a New Intelligence Algorithm

Author

Wang, Jingmin ; Ren, Guoqiao

Author_Institution

Dept. of Economy & Manage., North China Electr. Power Univ., Baoding

Volume

2

fYear

2006

fDate

17-19 July 2006

Firstpage

685

Lastpage

689

Abstract

An inverse problem of support vector machines (SVMs) was investigated. The inverse problem is how to split a given dataset into two clusters such that the margin between the two clusters attains the maximum. Here the margin is defined according to the separating hyper plane generated by support vectors. It is difficult to give an exact solution to this problem. An immunogenetic particle swarm incorporated intelligence algorithm was proposed to solve this problem. This study on the inverse problem of SVMs is motivated by designing a heuristic algorithm for generating decision trees with high generalization capability. The application in the recognition of the bank risk shows it is effective

Keywords

decision trees; generalisation (artificial intelligence); genetic algorithms; particle swarm optimisation; support vector machines; decision trees; generalization capability; genetic algorithm; heuristic algorithm; immunogenetic particle swarm; intelligence algorithm; inverse problem; support vector machines; Clustering algorithms; Decision trees; Entropy; Inverse problems; Kernel; Machine intelligence; Machine learning; Particle swarm optimization; Support vector machine classification; Support vector machines; genetic algorithm; incorporated intelligence algorithm; inverse problem; penalty factor; support vector machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Cognitive Informatics, 2006. ICCI 2006. 5th IEEE International Conference on

Conference_Location

Beijing

Print_ISBN

1-4244-0475-4

Type

conf

DOI

10.1109/COGINF.2006.365571

Filename

4216489