The invention belongs to the technical field of database privacy protection, and discloses a method for processing big data based on an MLDM algorithm meeting secondary aggregation. The method comprises the steps of introducing a k-means algorithm based on a (1,d,e)-MDAV algorithm; and when a big data set is processed, dividing the big data set into a plurality of small data sets firstly, then processing each small data set by using the (1,d,e)-MDAV algorithm, and finally combining the processed data, thereby enabling the whole data set to meet a (1,d,e)-diversity rule, and obtaining relatively low algorithm time complexity and relatively short algorithm time through improvement. According to the method, a k-means algorithm is introduced based on the (1,d,e)-MDAV algorithm; a new MLDM algorithm is proposed; and when the big data set is processed, the big data set is divided into the small data sets firstly, then each small data set is processed by using the (1,d,e)-MDAV algorithm, and finally the processed data are combined, so that the whole data set can meet the (1,d,e)-diversity rule, and the relatively low algorithm time complexity and the relatively short algorithm time can be obtained through the improvement.