Abstract: Aiming at the existing problems of multi-source information fusion, a method which simulates human thinking mechanism is presented.-In this method, rough sets theory and support vector machines are combined together, and by the aid of the strong qualitative analysis ability of rough sets theory and the quick association ability of support vector machines, multi-source information are fused. Genetic algorithm is introduced because of its excellent global optimization ability, and fused results are optimized. The results show that the proposed method has good abilities of-fault-tolerance, robustness and accuracy.