Abstract：
The study of terrorist organizations is of great significance to the analysis of the laws of the activities of terrorist organizations, and to effectively combat terrorist activities. As a new research method, data mining can be used to discover or find the hidden rules or knowledge in a large amount of data. In the past, the research of terrorist organizations is mostly from the political and religious point of view, there are few from the data level on the behavior of terrorist organizations to start with. With the help of global terrorism database GTD records from 1970 to 2014 since the rich terrorist incidents around the world, has a causal relationship in the time period before and after the assumption, proposed a causal modeling algorithm of sliding window FP-Growth based on attenuation of terrorist attacks, weapons, locations, attack targets behavior pattern mining. It firstly carries on the causal relationship modeling to the original data in time series, and then, on the basis of this, the historical event weights are attenuated, and finally, the FP-Growth is used to predict the behavior pattern of terrorist organizations. Through the contrast experiments of different sliding window attenuation factors, the accurate rate of attenuation factor is obtained, which has achieved good results, which has a certain prediction and guidance for the activities of terrorist organizations.
Tag：
点此返回栏目查看更多>>>参考论文
