Abstract: To deal with the mass alert data generated by heterogeneous network security equipments,this paper proposes a security data fusion algorithm based on weighted DS evidence theory and Fuzzy Cognitive Map.According to different weight and confidence of network security equipments,it initially fuses the alert data by DS evidence theory.Then it takes overall alert fusion through fuzzy reasoning of FCM.The proposed method overcomes the problem of high false negative rate and enhances the ability to cope with complex attacks.