An Intrusion Detection System Based on Hadoop
Zhiguo Shi, Jianwei An · 2015
In this paper, considering that the serious network security situation we are facing and the problem of an increasing amount of data generated by the network, we proposed an Intrusion Detection System based on Hadoop, due to the lack of the traditional K-Means algorithm exists at we propose an improved K-Means algorithm, we analyze the performance of the K-Means algorithm and the improved K-Means algorithm with KDD '99 data sets by using the Intrusion Detection System based on Hadoop. The experimental results show that the Accuracy Rate can reach 0.96, the Detection Rate can reach 0.89, the False Alarm rate the minimum is only 0.018. Three intrusion detection performance indicators are better than the traditional K-Means algorithm.