陆态网络敏感信息多级纵深过滤仿真研究  

Multi-level Deep Filtering Simulation Study of Terrestrial Network Sensitive Information

作　　者：张运香[1] 白建明[1] 金升灿[1] ZHANG Yun-xiang;BAI Jian-ming;JIN Sheng-can(Jiamusi University Business School,Jiamusi Heilongjiang 154007,China)

机构地区：[1]佳木斯大学商学院

出　　处：《计算机仿真》2019年第5期466-470,共5页Computer Simulation

摘　　要：为了保证陆态网络敏感信息多级纵深过滤的效率和时效性,针对当前方法敏感信息过滤效率低、过滤时间长的问题,提出基于ST-DFA的陆态网络敏感信息多级纵深过滤方法。对陆态网络信息做预处理,并利用陆态网络敏感信息的特点来构建网络敏感信息的语料库,检测语料库的初始和结束标记;在此基础上,利用非叶子节点等条件构建敏感信息决策树,能够获取敏感词的数量和敏感词的长度,通过对陆态网络信息流的分析,可以得到决策树的高度,通过对决策树的建立以及对敏感词的分析,将两者相结合,最终实现了陆态网络敏感信息多级纵深过滤。实验结果表明,提出方法在对陆态网络敏感信息多级纵深过滤时,具有较高的过滤效率,并且过滤时间较短,具有较好的时效性。In order to ensure the efficiency and timeliness of multistage depth filtering of sensitive information in Crustal Movement Observation Network of China,a multistage depth filtering method for sensitive information in Crustal Movement Observation Network of China based on ST-DFA was proposed.Through pretreatment of information in Crustal Movement Observation Network of China,the characteristics of sensitive information were used to construct the corpus of network sensitive information.Then,initial tag and ending tag of corpus were detected.On this basis,none-leaf nodes were used to construct the sensitive information decision tree,so that the number of sensitive words and the length of sensitive words could be obtained.By analyzing the information flow,the height of decision tree could be obtained.By establishing the decision tree and analyzing sensitive words,multistage depth filtering of sensitive information in Crustal Movement Observation Network of China was achieved.Simulation results show that the proposed method has higher filtering efficiency during the multistage depth filtering of sensitive information.Meanwhile,the filtering time is shorter,which has better timeliness.

关 键 词：陆态网络 敏感信息 多级纵深 滤波器 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]