The research by Google and the University of Texas at Austin that used data from NASA raised the prospects of new insights into the universe by feeding data into computer programs that can churn through information faster and more in-depth than humanly possibly, a technique known as machine learning.
In this case, software learned differences between planets and other objects by analyzing thousands of data points, achieving 96 percent accuracy, NASA said at a news conference.
The data came from the Kepler telescope which NASA launched into space in 2009 as part of a planet-finding mission that is expected to end next year as the spacecraft runs out of fuel.
The software’s artificial “neural network” combed through data about 670 stars, which led to the discovery of planets Kepler 80g and Kepler 90i. The latter, a scorching, rocky mass 30 percent larger than Earth, is the eighth planet found to be orbiting the same star.
Astronomers had never before observed an eight-planet network beside the solar system that includes Earth, researchers said.
“As the application of neutral networks to Kepler data matures, who knows what might be discovered,” said Jessie Dotson, a NASA project scientist for the Kepler space telescope. “I‘m on the edge of my seat.”
Christopher Shallue, an artificial intelligence researcher at Google, and Andrew Vanderburg, an astronomer at the University of Texas at Austin, said they plan to continue their work by analyzing Kepler data on more than 150,000 other stars.
Advancements in hardware and new techniques for machine learning have made it possible in recent years for automated software to tackle data analysis in science, finance and other industries.
Machine learning had not been applied to data acquired by the Kepler telescope until Shallue came up with the idea, he said.
“In my spare time, I started Googling for ‘finding exoplanets with large data sets’ and found out about the Kepler mission and the huge data set available,” he said. “Machine learning really shines in situations where there is so much data that humans can’t search it for themselves.”
Vanderburg has received funding through a NASA fellowship aimed at distant-planet researchers.
