基于混合神经网络的网络数据传输中恶意攻击数据辨识

Identification of Malicious Attack Data in Network Data Transmission Based on Hybrid Neural Network

传统的网络数据传输中恶意攻击数据辨识方法,只能经过一个训练周期,不能得到准确的恶意攻击数据辨识模型,导致正确分类恶意攻击数据文档数少,因此设计一种基于混合神经网络的网络数据传输中恶意攻击数据辨识方法。通过捕获目标网络数据并进行数据清洗和预处理,提取恶意攻击数据的特征参数。通过对网络数据进行多个训练周期,得到准确的恶意攻击数据辨识模型。辨别恶意攻击数据时,可以根据流量异常、行为异常和文件异常等元素与预定阈值进行比较。实验证明,该方法准确性高,具有研究价值。

the traditional malicious attack data identification method,only after a training cycle,cannot get accurate malicious attack data identification model,lead to the correct classification of malicious attack data document number,so design a kind of network based on hybrid neural net-work data transmission of malicious attack data identification mcthod.By capturing the targct network data and performing data cleaning and preprocessing,the characteristic parameters of the malicious attack data are extracted.Through multiple training cycles on the network data,an accurate malicious attack data identification model is obtained.When identifying malicious attack data,you can compare with predetermined thresholds based on elements such as traffic,behavior and file exceptions.The iments proved that this method has high accuracy and research value.