论文部分内容阅读
针对大数据环境下互联网安全日趋严重的问题,构建了一个基于粗糙集和改进新种群生成方式的遗传算法优化BP网络的网络攻击检测模型.该模型首先通过粗糙集进行除噪降维,然后通过选择部分最优父代个体进行交叉变异后替换父代最差个体的方式生成新的种群,同时改进新种群生成方式的遗传算法可以为BP网络提供更合理的初始参数,从而解决了BP网络中存在的检测速率慢且容易陷入局部最优等问题.最后通过仿真实验验证了提出的模型检测正确率高,且能够缩短检测时间.“,”In view of the situation that Internet security is becoming increasingly serious and important in the big data environment, a network attack detection method is developed which optimizes the BP network by using the rough set and improved genetic algorithm of a new population generation model. Based on the improved genetic algorithm, the new population generation model produces a new population by selection of the optimal parent individuals undertaking crossover mutation to replace the worst parent individuals. The improved genetic algorithm of the new population generation model provides the reasonable parameters to BP network and supplies a better solution to the slow detection rate and local optimum problem of BP network. Simulation tests are provided to verify the improvement of detection accuracy and reduction of detection time of this model.