<p>In the last few decades, astronomers have discovered thousands of exoplanets worldwide but little is known about how many support life. Scientists in India have taken a jab at the problem by devising a new approach that could help strip down the mystery of many of these worlds.</p>.<p>Astronomers of the Indian Institute of astrophysics (IIA) in Bengaluru and BITS Pilani (Goa campus) have devised a new approach, called an anomaly detection method through which they can identify potentially habitable planets with higher probability.</p>.<p>Dr Margarita Safonova, a visiting astronomer at IIA, explained that the new method was developed because there “are as many more planets than there are stars in the galaxy and because the number of potentially habitable exoplanets is rising every day.”</p>.<p>Out of 8,000 exoplanets identified, about 4,500 are seen as potentially habitable, she said, adding that assessments by Planetary Habitability Laboratory in Arecibo have determined that about 60 are potentially habitable through various parameters such as planetary radius and appropriate distance from the sun (“Goldilocks” habitable zone).</p>.<p>However, a colleague at Bits Pilani (Goa Campus), proposed a novel new method that postulates looking for new worlds on the assumption that the Earth itself is an “anomaly.”</p>.<p>Associate professor Dr Snehanshu Saha of the Anuradha and Prashant Palakurthi Centre for Artificial Intelligence Research (APPCAIR) at Bits Pilani explained that his new method operates on the postulation that the Earth is an anomalous world because it is the only planet that we know of that can support life.</p>.<p>“Our planet has a unique set of physical properties or ‘features,’ which a machine learning tool can leverage to look for in other worlds,” he said. He added that a new AI system, named Multi-Stage Memetic Binary Tree Anomaly Identifier (MSMBTAI), was created to conduct the search using this as a parameter.</p>.<p>“The AI tool will scan international databases holding information on all known exoplanets. It will evaluate habitability perspectives across 4,800 data points and 64 planetary features,” Dr Saha said.</p>.<p>The study has been published in the journal Monthly Notices of the Royal Astronomical Society (MNRAS).</p>.<p>Dr Safonova said that AI is important because, amid ever-increasing numbers of exoplanets being found, it is not possible to scan every new planet to identify worlds that are potentially similar to Earth.</p>.<p>The study has already identified a few planets which exhibit similar anomalous characteristics as Earth via the proposed technique, “which shows reasonably good results, in agreement with what astronomers believe,” a statement from the researchers said.</p>.<p>The new method has also resulted in similar results in terms of anomalous candidate detection when it did not use surface temperature as a feature, compared to when it actually did. Scientists expect this to make the future analysis of exoplanets easier.</p>.<p><strong>Watch latest videos by DH here:</strong></p>
<p>In the last few decades, astronomers have discovered thousands of exoplanets worldwide but little is known about how many support life. Scientists in India have taken a jab at the problem by devising a new approach that could help strip down the mystery of many of these worlds.</p>.<p>Astronomers of the Indian Institute of astrophysics (IIA) in Bengaluru and BITS Pilani (Goa campus) have devised a new approach, called an anomaly detection method through which they can identify potentially habitable planets with higher probability.</p>.<p>Dr Margarita Safonova, a visiting astronomer at IIA, explained that the new method was developed because there “are as many more planets than there are stars in the galaxy and because the number of potentially habitable exoplanets is rising every day.”</p>.<p>Out of 8,000 exoplanets identified, about 4,500 are seen as potentially habitable, she said, adding that assessments by Planetary Habitability Laboratory in Arecibo have determined that about 60 are potentially habitable through various parameters such as planetary radius and appropriate distance from the sun (“Goldilocks” habitable zone).</p>.<p>However, a colleague at Bits Pilani (Goa Campus), proposed a novel new method that postulates looking for new worlds on the assumption that the Earth itself is an “anomaly.”</p>.<p>Associate professor Dr Snehanshu Saha of the Anuradha and Prashant Palakurthi Centre for Artificial Intelligence Research (APPCAIR) at Bits Pilani explained that his new method operates on the postulation that the Earth is an anomalous world because it is the only planet that we know of that can support life.</p>.<p>“Our planet has a unique set of physical properties or ‘features,’ which a machine learning tool can leverage to look for in other worlds,” he said. He added that a new AI system, named Multi-Stage Memetic Binary Tree Anomaly Identifier (MSMBTAI), was created to conduct the search using this as a parameter.</p>.<p>“The AI tool will scan international databases holding information on all known exoplanets. It will evaluate habitability perspectives across 4,800 data points and 64 planetary features,” Dr Saha said.</p>.<p>The study has been published in the journal Monthly Notices of the Royal Astronomical Society (MNRAS).</p>.<p>Dr Safonova said that AI is important because, amid ever-increasing numbers of exoplanets being found, it is not possible to scan every new planet to identify worlds that are potentially similar to Earth.</p>.<p>The study has already identified a few planets which exhibit similar anomalous characteristics as Earth via the proposed technique, “which shows reasonably good results, in agreement with what astronomers believe,” a statement from the researchers said.</p>.<p>The new method has also resulted in similar results in terms of anomalous candidate detection when it did not use surface temperature as a feature, compared to when it actually did. Scientists expect this to make the future analysis of exoplanets easier.</p>.<p><strong>Watch latest videos by DH here:</strong></p>