Prediction of terrorist attacks based on GA-BP neural network
Qinghao Li, Zonghua Zhang, Zhen Shen · IOP Conference Series Materials Science and Engineering · 2019
Terrorism is a common threat to mankind. Combating terrorism is also the responsibility that every country should assume. In the face of the development trend of terrorism, in-depth quantitative analysis of data related to terrorist attacks will help deepen people's understanding of terrorism and provide valuable information support for counter-terrorism, and effectively improve the pertinence and efficiency of the anti-terrorism struggle. This paper designs a risk assessment and prediction system for terrorist attacks. The mathematical model calculates the relative risk index of each type of target by factor analysis. The evaluation indicators include three factors: "threat", "vulnerability" and "consequence". Mathematical model uses neural network to evaluate and predict risk index. However, because the BP neural network is easy to fall into the local best, it is difficult to jump out. Therefore, the genetic algorithm (GA) optimizes the initial weight threshold of the BP neural network to enhance the prediction accuracy. Finally, the simulation experiments of 21 main targets are carried out to verify the effectiveness of the model, so as to carry out accurate strategy analysis.