A deep learning framework for predicting cyber attacks rates

EURASIP Journal on Information Security

Peng Zhao¹  Maochao Xu²  Xing Fang³  Shouhuai Xu⁴ 

[1] 0000 0000 9698 6425, grid.411857.e, Department of Computer Science, Jiangsu Normal University, 221110, Xuzhou, China;0000 0004 1936 8825, grid.257310.2, Department of Mathematics, Illinois State University, 61761, Normal, IL, USA;0000 0004 1936 8825, grid.257310.2, School of Information Technology, Illinois State University, 61761, Normal, IL, USA;0000000121845633, grid.215352.2, Department of Computer Science, University of Texas at San Antonio, 78249, San Antonio, TX, USA;

关键词: ARIMA;    GARCH;    RNN;    Hybrid models;    LSTM;    Deep learning;    BRNN-LSTM;   

DOI  :  10.1186/s13635-019-0090-6

来源: publisher

【 摘 要 】

Like how useful weather forecasting is, the capability of forecasting or predicting cyber threats can never be overestimated. Previous investigations show that cyber attack data exhibits interesting phenomena, such as long-range dependence and high nonlinearity, which impose a particular challenge on modeling and predicting cyber attack rates. Deviating from the statistical approach that is utilized in the literature, in this paper we develop a deep learning framework by utilizing the bi-directional recurrent neural networks with long short-term memory, dubbed BRNN-LSTM. Empirical study shows that BRNN-LSTM achieves a significantly higher prediction accuracy when compared with the statistical approach.

【 授权许可】

CC BY