关键词: 大数据, 网络信息安全, 网络攻击, 文本分类, 朴素贝叶斯分类法, 改进权重朴素贝叶斯分类方法

Abstract: In recent years, with the rapid development of information technology in the era of big data, information security research has also been rapid development, more and more network information security attacks continue to occur and are reported. In order to protect the information security of the network, it is very important to establish the network information security attack and tracking system. Network information security attack and tracking analysis system is based on network attacks, which can leave a large amount of network attack traces, and the advantages of large data analysis platform for multi-source massive data analysis, multi dimension and multi angle analysis. Which is a very important technology in the network information security attack and tracking analysis system based on large data. This paper chooses Naive Bayesian as a text classification algorithm, because the characteristics of the Naive Bayesian classification algorithm is generally very difficult to meet in the reality, in order to relax the assumption in a certain degree, this paper proposes a classification method based on feature items improved weight Naive Bayesian, which is based on the improved chi square statistical feature selection method and weighted Naive Bayesian classification algorithm. The experimental results show that the classification method based on the feature of the improved weight of the Naive Bayesian classification method has a certain improvement compared with the classification results.

Key words: Big Data, Network Information Security, Network Attack, Text Classification, Native Bayes Classifier, Improve the method of weighted naive bayes classification