Big data intrusion detection method based on naive Bayesian model and cloud security

• Main Authors: ZOU JUNGUO, WEI GUANGXING, DAI YUE, LI HUA, MIAO MENGJUN, CHEN YINYAN
• Format: Patent
• Language: Chinese; English
• Published: 01.06.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a big data intrusion detection method based on a naive Bayes model and cloud security, and the method comprises an intrusion detection method based on naive Bayes and cloud security. The intrusion detection method based on naive Bayes and cloud security mainly comprises five parts, namely a cloud data security module, a data processing module, an attribute weight determination module, an NB model and an intrusion monitoring system, wherein the data processing module is divided into two parts, namely an improved attribute selection algorithm and data discretization. The attribute weight determination module uses a fisher score value to measure a weighting coefficient, and the NB model is mainly hidden naive Bayes. According to the method, the naive Bayes model is optimized from the aspect of an attribute selection algorithm, the structure of the naive Bayesian model is improved by using weights, and the improved attribute selection algorithm and the weighted hidden naive Bayesian model