Purpose: discuss the role of the neural network (NN) theory in the computer network security evaluation.
Method: propose three-level and four-class indicator system suitable for network security evaluation, establish the
network security evaluation system model based on NN, optimize the NN model by using the particle swarm, collect 100-
group data on the computer network security evaluation of different scales via expert scoring, and normalize them; Result:
the evaluation model based on NN is simple and practicable to network security evaluation and can eliminate disturbance
of the subjective factors of the human being. The simulation results indicate that the system can reduce relative output
error and improve correctness rate of evaluation. Conclusion: The NN model is very valuable in research on the computer
network security evaluation system, which can offset weaknesses of the past evaluation methods to some extent, improve
precision of the evaluation results, and provide reference to prediction and control of the network security problems in
future.
Computer Network Security Evaluation Based on LM-BP Neural Network
