Prediction of solar flares is important for preparedness for the effects of space weather, including a direct impact on living organisms. This is reported by a group of scientists from PNAS.
A flares prediction model using machine learning, created by American scientists, analyzes their maximum number during the day. This system, first of all, will contribute to a more accurate determination of the temperature conditions on the Sun. Machine learning is used to develop algorithms that can explore large amounts of data and, as a result, make decisions.
Read more on the topic:
During the creation, more than a thousand photographs of the Sun for the period from 2010 to 2016 were analyzed. AI rather quickly learned to distinguish quiet zones on the Sun from potentially dangerous ones. The accuracy of its forecasts is 75 – 92%. In its predictions, AI relies on changes in the electrical activity of the solar photosphere in places where flares were observed.
Then, with the help of subsequent analysis of data obtained through solar satellites, scientists will be able to figure out which factors have a significant impact on the state of solar matter and how to predict such disasters with 100% accuracy.
We want to remind you:
Subscribe to The Coin Shark news in Facebook: https://www.facebook.com/coinshark/