基于大数据分析恐怖袭击风险预测研究与仿真

Research and Simulation of Terrorist Attack Risk Prediction Based on Big Data Analysis

在对恐怖袭击风险预测过程中,在建立预测模型时,由于受到恐怖袭击带有伪装性的影响,存在大量的伪装性样本和干扰性数据,真实的样本不充足、可统计数据波动较大。导致预测过程很容易受到干扰,存在预测精度低的问题。提出了大数据分析的恐怖袭击风险预测方法。建立恐怖袭击风险综合评判的大数据分析模型,采用大数据分析模型对恐怖袭击历史数据中隐含的可演化信息进行学习,利用所获取的结果进行未来的恐怖袭击预测。预测过程中融入多步时间序列预测中的递推计算的思想,将每一步预测的不确定性作为下一次预测迭代的输入要素加以充分考虑,提升预测精确度。仿真结果证明,采用大数据分析的恐怖袭击风险预测方法精确度和效率都比较高。

A risk prediction method of terrorist attack is presented based on big data analysis. A data analysis model of comprehensive evaluation for major terrorist attack risk is established, the implied information in the historical data is studied by using big data analysis model to the terrorist attacks, and the future terrorist attacks are predicted by using the obtained results. Recursive calculation of multi - step time sequence is applied to the prediction process, every step will predict the uncertainty as a factor to give full consideration to the input of the next prediction iteration, and the prediction accuracy is improved. The simulation result proves that theproposed method has high precision and efficiency.