Applying this process, researchers Christopher Shallue and Andrew Vanderburg trained a computer to identify Kepler's light readings, that is the minute changes in brightness that are caused by a planet passing in front of a star.
The system, known as Kepler-16b, is roughly 200 light years from Earth. Similarly, the larger and more gaseous planets in the Kepler-90 system are farther away from the parent star, quite similar to the relation between Jupiter, Saturn, Neptune, and Uranus with the sun.
NASA have admitted many of these stars contain habitable planets - planets where the conditions are right for water to exist - but the agency does not know yet which of these could sustain life.
While machine learning has been used before in the search for planets beyond our solar system, it's believed to be the first time an artificial neural network like this has been used to find a new world.
In order to locate Kepler-90's planets NASA's AI had to scan through a daunting thirty five thousand potential signals from distant stars, over a period of four years.
The Trappist-1 star system, which hosts a record seven Earth-like planets, was one of the biggest discoveries of 2017.
U.S. judge to lift house arrest for former Trump campaign manager Manafort
Manafort can't go near any airports, train stations or bus stations unless he gets approval for travel from the judge in advance.
Full text: Republicans unveil their final tax bill
While GOP lawmakers aimed to make the cuts permanent, the individual rate deduction is slated to expire in 2025. Thus, he concludes that, on balance, the nation is better off with the bill than without it.
National study: Adolescent marijuana use edges upward
For eighth-graders, the percentage who had used illicit drugs in the previous year was just 12.9%, while 18.2% had had a drink. Teens today are using fewer drugs than the age group has used over the the past 43 years, with one clear exception: marijuana.
The latter, a scorching, rocky mass 30 per cent larger than Earth, is the eighth planet found to be orbiting the same star.
Shallue, a senior researcher with Google AI, became excited about the prospects for machine learning in space exploration when he realised just how much data Kepler had gathered over the past four years.
This kind of monitoring had previously been done by using automated tests or the human eye but, as Shallue explained, the new method was able to capture some of the weakest signals that had been previously missed. The collected data is then analysed by Google's machine learning AI, which is how the new discovery was made. They plan to follow up these discoveries by using AI to examine the entire Kepler system, which is comprised of 150,000 stars. The astronomers trained the programme on a set of about 15,000 signals, and it identified planets correctly.
"What is perhaps most exciting is that they are able to find planets that were previously missed, suggesting there are more yet to be found using this approach", Suzanne Aigrain, an astrophysicist at Oxford University said to The Guardian.
Kepler found a planet that orbits two stars, known as a binary star system, in 2011. For instance, in the solar system itself, the outer planets formed in the solar system's cooler part because this is where ice can remain solid and clump together to make more massive planets. He also compared the finding of exoplanets to sifting through rocks with a fine sieve to find jewels. Kepler-90i was not the only jewel this neural network sifted out.
One planet in particular, Kepler-90i, has attracted the attention of astronomy boffins for bringing the total Kepler planet count to eight - just like our solar system.