Research on Prediction Method of Terrorist Attack Based on Random Subspace

In the process of risk prediction of terrorist attacks, the prediction model is set up, due to the impact of terrorist attacks with camouflage, there are a large number of camouflage sample and intrusive data, real samples are not enough, so the statistics can be volatile. It leads to the prediction process quite easily disturbed, and exists the problem of low accuracy. This paper puts forward the major terrorist attack risk prediction methods of data analysis, establishes a comprehensive evaluation of major terrorist attack risk data analysis model, uses big data analysis model to study the terrorist attacks of the evolution of information implied in the historical data, the results obtained are used to predict the future terrorist attacks. The simulation results show that terrorism risk prediction method based on random subspace has high precision and efficiency.