基于RNN的网络安全态势预测方法

Prediction of Network Security Situation Based on RNN

从宏观角度把控网络安全状况并预测态势的变化趋势是当前网络安全领域的热点,网络安全态势预测技术是其中最关键的一环,但目前主流预测方法存在精度低、实时性差等问题。循环神经网络RNN目前广泛应用于自然语言处理领域并取得了显著成果,其主要优越性在于处理序列数据。因此,利用网络态势预测即时间序列预测的特点,提出一种基于RNN的网络安全态势预测方法,并通过仿真实验验证该方法的可行性和准确性。

It is a hot research field of network security to contrul the network security situation and predict the trend of the situation from macroscopie view. The network situation prediction method is the most critical part, hut the mainstream method has problem with low precision or real-time prediction. Recurrent Neural Network is popular model that has shown great promise in many NLP tasks. The key idea behind RNN is to make use of sequential information. Proposes a new prediction method based on RNN considering the data of network situation is a data of time series and experimental results show that this method is feasihle and precise.