With the fast development of the technology in Database and Internet, excessive information has become a problem that decision makers engage in electronic commerce have to affront. This paper firstly establishes the decision table for excessive data, then reduces knowledge of the decision table, and obtains the best rules. This can reduce the redundancy of the data, and improve the efficiency of knowledge obtaining. Because of the high time complexity and spatial complexity the general algorithm has, this paper proposed a new algorithm based on analogical matrix, and carried out the algorithm. The practice proves that the algorithm can improve the efficiency of knowledge obtaining, and the rules is tidy and best.