Just how many benefits can artificial intelligence actually bring? Amid scepticism and doubts, in some cases there is even the prospect of saving lives. Using artificial intelligence to predict when aftershocks will occur and how strong they will be after an earthquake is a new application, as emerges from an article in ‘Nature’ in September.
This system works on the basis of machine learning: a branch of artificial intelligence that develops models and algorithms not from instructions given by an operator, but from a flood of data coming, for example, from the internet or from databases.
In this case, data on earthquakes are used: their location, magnitude, duration and so on, in short all the characteristics of a tremor and also of the subsequent aftershocks, their number, strength, duration and so on. In this way, with such a vast amount of data at hand, when an earthquake strikes the AI can predict the aftershocks on the basis of a statistical model.