A network defense system must balance both security and stability. To enhance its risk control capabilities， a comprehensive risk assessment model for the network defense system based on the combination of game theory and weight assignment is proposed. The risk evaluation index system is established through literature analysis， expert surveys， and factor analysis. The subjective and objective weights of the evaluation indices are calculated using the continuous ordered weighted average （COWA） and the entropy weight method （EWM）. The optimal combination weights are determined using game theory， and the risk evaluation level of the network defense system is assessed using the fuzzy matter-element method. A case study of a local area network （LAN） project validates the model. The results demonstrate that the model is a useful reference for assessing the risk of network defense systems.

Attack and defense are dialectical unity， and their mutual transformation is a direct embodiment of the combat process. In network attack and defense， the offense usually dominates while the defense is in a relatively weak position， and the idea of dynamic defense is valued. By combing and analyzing the dynamic defense of cyberspace， the factors affecting the defense capability of cyberspace are obtained， and a set of cyberspace defense index systems based on dynamic empowerment is constructed. To eliminate the interference of human preference on the results， the entropy weight method is adopted to simplify the index set， and the comprehensive weight of each index is weighted. The TOPSIS（technique for order preference by similarity to an ideal solution）-gray correlation analysis method is applied to construct an evaluation model to realize the evaluation of cyberspace defense capability， and the effectiveness and feasibility of the evaluation model are proved by specific cases.