基于 C NN 和 LSTM 深度网络的伪装用户入侵检测
*
王 毅
1
,冯小年
2
,钱铁云
1+
,朱 辉
3
,周 静
3
1. 武汉大学 计算机学院,武汉 430072
2. 中国电力财务有限公司,北京 100005
3. 北京汇通金财信息科技有限公司,北京 100094
CNN and LSTM Deep Network Based Intrusion Detection for M alicious Users
WANG Yi
1
, FENG Xiaonian
2
, QI AN Tieyun
1+
, ZHU H ui
3
, ZHOU Jing
3
1. School of Computer Science, Wuha n University, Wuhan 430072, China
2. China Power Finance Co ., Ltd., Beijing 100005, China
3. Beijing Huitong Financial Information Technology Co., L td., Beijing 1 00094, China
+ Corresponding author: E-ma il: qty@whu.edu.cn
WANG Yi, FENG Xiaonian, QIAN Tieyun, et al. CNN and LSTM d eep network based intrusion de tection for
malicious users. Journal of Frontiers of Computer Science an d Technology, 2018, 12(4):575-585.
Abstract: The intrusion detection of internal malicious users, as an active securi ty protection technology, has been a
hot rese arch topic in recent years. Existing methods are unable to accurat ely model the users be havior. Thi s paper
proposes a no vel CCNN-LSTM met hod which combines the convolution neural network (CNN) and lon g short-term
me mory (LSTM) neural network for camouflage intrusion detection. The basic idea is to use convolution neural net-
wo rk to capture the loca l correlation in users activity data, and use l ong short-term memory neural network to deal
wit h sequential relationship and long-range dependency. The proposed method can automatically learn the represen-
tation of data without artificial extraction of complex features, and can also scale to large volume of high dimensional
data. The experimental results show that the proposed me thod has hig her detecti on rate and lower detection co st
than a number of baselines.
Ke y wo rds: intrusion detection of malicious users; depth neural network; convolution neural network; long and
short-term memory a rtificial neura l network
* The National Natural Science Foundation of China under Grant No. 61572376 (国家自然科学基金).
Received 2017-07, Accepted 2017-11.
CN KI 网络出版: 2017-11-28, http://kns.cnki.net/kcms/d etail/11.5602. TP.20171128.0823.002.html
ISS N 1673-9418 CO DEN JKYTA8
Journal of Front iers of Computer Science and Technology
1673-9418/2018/12(04)-0575-11
doi: 10.3778/j.issn. 1673 -9418.1707049
E-mail: fcst@vip.1 63.com
http://www.ceaj.o rg
Tel: +86-10-89056056
评论0
最新资源