摘要
网络安全状态数据具有数据量大、特征数目繁多以及连续型属性多等特点.态势预测问题可转化为海量数据的预测问题.以网络安全态势研究为应用背景,提出了一种基于改进的粒子群优化算法来优化反向传播神经网络的态势预测模型.利用IPSO内在的隐并行性和很好的全局寻优能力对BP网络的权值和阈值进行优化并建立预测模型对网络安全态势进行预测.仿真实验证明其改善了传统BP网络在预测应用中的不足,有效提高了态势预测的精准度.
Network security status data has volume, variety, number of features and continuous multi - at- tribute characteristics. However, situation prediction can be transformed into massive data prediction. Un- der the background of network security situation application research, a back-propagation neural network (BPNN) situation prediction model optimized by improved particle swarm optimization (IPSO) algorithm. With the inherent and implicit parallelism and good global optimization ability of IPSO, weights and thresh- olds of BPNN can be optimized, a predictive model is built to predict the network security situation. Simu- lation results sho-s that the algorithm proposed in this paper can effectively improve the traditional BP neu- ral network deficiencies in predicting application, and the situation prediction accuracy.
出处
《闽江学院学报》
2013年第5期78-83,共6页
Journal of Minjiang University
关键词
网络安全
态势感知
粒子群优化算法
反向传播神经网络
态势预测
network security
situation awareness
particle swarm optimization
back propagation neuralnetwork
situation prediction