AI found, volunteers sorted: 8,000 new "eclipsed double stars"

Volunteers from around the world participating in NASA 's exoplanet research satellite project have scanned the sky and identified a total of 10,001 "eclipsing binary star" systems, 7,936 of which were previously unknown.

According to the Washington Post, AI-powered algorithms initially identified approximately 1.2 million candidate systems.

These candidates were then narrowed down to about 60,000 priority targets and submitted for review by citizen scientists.

Ultimately, 1 in 10,000 of these observations were confirmed to be true eclipsing binary stars.

WHAT IS AN "ECLIPSE DOUBLE STAR"?

These systems are called "eclipsing binaries" because as the two stars orbit each other, one will occasionally pass in front of the other, causing a noticeable dip in the star's brightness.

This temporary decrease in brightness allows astronomers to more accurately understand the size, mass, and evolution of stars.

" ARTIFICIAL INTELLIGENCE IS NOT ENOUGH"

"AI can scan big data quickly, but it can miss some complex patterns. That's why we need the intuition of the human eye," said Veselin Kostov (NASA Goddard Space Flight Center and SETI Institute), one of the project leaders.

Kostov emphasized that these newly discovered stars are of great importance not only in terms of stellar physics, but also in terms of the discovery of potential exoplanets.

The project is still ongoing and anyone can participate. Anyone with an internet connection can join the project website and analyze the data.