A Multi-Source Information Fusion Method Based on Rough Set Theory and Support Vector Machine and Its Application
Huang Ku · 2005
Aiming at the existing problem of multi--source information fusion, a method which simulates human thinking to realize information fusion is presented in this paper. In this method, multi--source input information is reduced firstly by using the strong qualitative analysis ability of rough set theory, and the noise and redundancy in the samples are removed. On the basis of it, using support vector machine, reduced information is fused. In order to obtain optimum fusing accuracy, genetic algorithm is used to optimize fusion parameters. Concrete example of text recognition shows that the proposed method has fine fault--tolerance, robustness and accuracy.