✅ Fact-checked    

China has learned how to predict terrorist attacks

Mathematicians at Zhejiang University have developed a machine learning system that, according to messages on the Internet and other open data can determine in advance the approximate time and place of the attack. Their development was published by the scientific journal Science Advances.

“Using only publicly available data, we have created a model capable of predicting many terrorist attacks a week in advance in regions of the world where terrorists are particularly active,” the study said.

The Chinese mathematicians, led by Professor Andre Piton, combined data on terrorist attacks that occurred in all countries around the world between 2003 and 2016. After analyzing them using various machine-learning algorithms, they tried to find indicators that could be used to predict further attacks by extremists. Their calculations showed that such predictions are only possible in those parts of the world where terrorist attacks occur quite frequently. Using the algorithm there, it would be possible to predict about half of the incidents a week before an attack.

They have recommended their development to the governments of countries where terrorist attacks are particularly frequent, to increase the chances of preventing them and to reduce the resources devoted to dealing with false threats.

In the process, the Chinese scientists also concluded that extremists are most active in economically developed regions with low population densities and good road coverage.

After the defeat of the Caliphate in Syria and Iraq, terrorism in Southeast Asia gained new momentum. A modern phenomenon is that radical Islam and jihadism are now taking over “new beachheads” without resorting to overt terror. They support their candidates going into politics, manipulate public opinion, and influence society through affiliated organizations.

Views: 411


Add a comment!

Your name:
Your Email:
  • bowtiesmilelaughingblushsmileyrelaxedsmirk
The code:
Кликните на изображение чтобы обновить код, если он неразборчив
Enter a code: