Threat Intelligence Disclosure Trend Analysis Model Based on Time Series

Threat Intelligence Disclosure Trend Analysis Model Based on Time Series

연구 분야: Safety

학회: 2021 2nd International Conference on Electronics, Communications and Information Technology (CECIT)

In recent years, the number of malicious attacks on the Internet using threat intelligence information has increased dramatically. It is necessary to establish a scientific threat intelligence trend prediction model, predict and analyze the time trend of threat intelligence, and provide decision support for the formulation of targeted network security protection strategies. In order to achieve more accurate network security measures, integrated with Autoregressive Integrated Moving Average (ARIMA) model and Generalized AutoRegressive Conditional Heteroskedasticity (GARCH) model, this paper proposes a selection algorithm based on normal testing. Under the premise of retaining data sequence characteristics, according to data, more normal tests are provided to make data more accurate predictive threat information trends. A predictive model based on selection algorithms is proposed to predict the time trend of threat intelligence and enhance the dimension of network security defense. Experiments have proved that the prediction model based on the selection algorithm has higher accuracy and real-time performance in threat intelligence data prediction.