语义分析及向量化大数据XSS入侵识别

Identifying of Big Data XSS Intrusion with Semantic Analysis and Vectorization

通过语义情景分析及向量化对访问流量语料库大数据进行词向量化处理,实现了面向大数据XSS入侵智能检测研究.利用自然语言处理方法进行数据获取,数据清洗,数据抽样,特征提取等数据预处理;设计了基于神经网络的词向量化算法,实现了词向量化得到词向量大数据;通过理论分析和推导,实现了不同深度的深层神经网络智能检测算法;设计不同的超参数并进行反复的实验,分别得到最大识别率、最低识别率、识别率均值、方差、标准差、损失误差变化过程曲线图和词向量样本余弦距离变化曲线图等结果,证明了研究的语义情景分析及向量化面向大数据XSS入侵智能检测系统具有识别率高,稳定性好,总体性能优良等优点.

First of all,the access traffic corpus big data is processed with word vectorization through the methods of semantic scenario analysis and vectorization,and then the research of intelligent detection oriented to big data XSS intrusion is realized.First,it makes the methods of natural language processing for data acquisition,data cleaning,data sampling,feature extraction and so on data pretreatment;Second,the algorithm of word vectorization based on neural network is designed,and to realize word vectorization and get big data of word vectorization;Third,through theoretical analysis and deductions,the intelligent detection algorithms of varieties of deep neural networks with different layers are realized;Finally,to design different hyper-parameters and make repeated experiments,respectively get the highest recognition rate,the minimum recognition rate,average recognition rate,variance,the standard deviation,the curve diagram of recognition rates change,the curve diagram of error of loss change and the curve diagram of cosine proximity change of word vector samples etc.As a result,which proved the method researched for intelligent detection oriented to big data XSS intrusion through the methods of semantic scenario analysis and vectorization has high recognition rate,good stability,the excellent overall performance and so on.