Eight additional potential indications of extraterrestrial intelligence have been discovered by a machine learning algorithm.
The programme was taught to analyze huge datasets from radio
telescopes, specifically looking for signals that could not have a terrestrial
origin. It was created by Peter Xiangyuan Ma of the University of Toronto in
Canada along with his team.
Massive data sets have been combed through as part of the
Breakthrough Listen Initiative and the Search for Extraterrestrial Intelligence
(SETI) effort to hunt for indications of technologically advanced civilizations
elsewhere in the cosmos. Filtering away the tens of millions of false positives
that can be found in these databases is the difficult part. It is now simpler
to detect potential signals of interest thanks to the machine learning
algorithm created by Xiangyuan Ma and his colleagues, which has been trained to
recognize distinctive patterns in the data.
The group examined 480 hours of observations of 820 stars
made by the West Virginia-based Robert C. Byrd Green Bank Telescope using the
algorithm. The programme found nearly 3 million unique patterns, which were
further reduced to only 20,515 by the algorithm. The team found eight promising
signs of interest upon ocular assessment.
Five stars that are between 30 and 90 light-years away from
Earth are the source of these emissions. HIP 62207, one of these stars,
resembles the sun. The signals, however, did not last over time and vanished
when the stars were studied again. The team nevertheless recommends additional
observations of these targets.
The machine learning-based techno signature search
represented by this work is the most thorough one to date. With the help of
automated assessments like this one, future research will be easier to organise
and our chances of finding alien civilizations may increase. Even though it's
unclear where these signals came from, their discovery offers fresh insights
into how the cosmos works.