Network intrusion detection has always been a hot topic in the field of computer network security .Currently, the network is faced with many security risks.Support vector machine (SVM) was chosen as the classification model of machine learning and the pa-rameters of SVM were optimized by the improved quantum particle swarm optimization (IQPSO) in order to improve the accuracy of network intrusion detection and detection efficiency .Based on this , a network intrusion detection method based on IQPSO-SVM algo-rithm was designed .The experimental results show that IQPSO-SVM algorithm not only has obvious improvement in efficiency , but also increases the accuracy of network intrusion detection by 4.26%and 7.12%, compared with that by using Deep neural network (DNN) and SVM algorithms,respectively;meanwhile ,the false positive rate was reduced by 0.93%and 3.31%, respectively, and the missing rate was decreased by 0.52%and 1.26%, respectively.%网络入侵检测一直是计算机网络安全领域的研究热点,当前网络面临着诸多的安全隐患.为了提高网络入侵检测的准确性与检测效率,首先选择支持向量机(SVM)作为机器学习分类模型,其次用改进的量子粒子群算法(IQPSO)对支持向量机的参数进行优化.以此为基础,设计了一种基于IQPSO-SVM算法的网络入侵检测方法.实验结果表明,IQPSO-SVM算法相对于深度神经网络(DNN)算法和SVM算法,不仅在效率上有了明显的改善,而且在网络入侵检测的正确率上分别提高了4.26%和7.12%,在误报率上分别降低了0.93%和3.31%,在漏报率上分别降低了0.52%和1.26%.
展开▼
