基于隐私大数据的网络信息防泄漏推荐算法  

A Recommendation Algorithm for Network Information Leakage Prevention Based on Private Big Data

作　　者：黄卫[1] 江官星[1] HUANG Wei;JIANG Guan-xing(College of Science and Technology,Nanchang Hangkong University,Gongqingcheng Jiangxi 332020,China)

机构地区：[1]南昌航空大学科技学院,江西共青城332020

出　　处：《计算机仿真》2022年第11期483-486,500,共5页Computer Simulation

基　　金：江西省教育厅科学技术研究项目(191615)。

摘　　要：当前网络信息防泄漏推荐算法忽略了对网络大数据的分类和保护,其应用结果存在推荐误差较高、覆盖率偏低的问题。为解决上述问题,提出基于隐私大数据的网络信息防泄漏推荐算法。利用转换随机化方法完成大数据的转换,得出大数据特征,对隐私大数据实现保护。利用协同滤过方法实现数据项目特征的划分和评价,根据群组和评价分数推荐兴趣点,完成网络信息防泄漏推荐。实验结果表明,所提算法的算法的推荐精度较高,覆盖率较高,且F1值始终保持较好水平。以上实验结果说明所提算法具有较理想的实用性。The current network information leakage prevention recommendation algorithm ignores the classification and protection of network big data,and its application results have the problems of high recommendation error and low coverage.In order to solve the above problems,a recommendation algorithm for network information leakage prevention based on private big data was proposed.At first,the transformation and randomization were applied in the transformation of big data.Then,the feature of big data was obtained to protect the privacy of big data.Moreover,the collaborative filtering method was used to classify and evaluate the characteristics of data items.Finally,interest points were recommended by groups and evaluation scores.Thus,the recommendation for network information leakage prevention was completed.Experimental results show that the proposed algorithm has high recommendation accuracy and coverage rate,and the F1 value always keeps a high level.In addition,this algorithm has good practicability.

关 键 词：隐私数据 推荐算法 保护处理 协同滤过 兴趣点 

分 类 号：TM76[电气工程—电力系统及自动化]