Data Science and Machine Learning Key to Enhanced Insight into Supermassive Black Hole

 The study and related understanding of cosmos, galaxies and beyond have been tantalizing to humans for eons. As technology and computing powers improved, especially exponentially over the past two decades, the human reach to comprehend and grasp the distant astronomical objects in our universe has become less difficult too. With the help of James Webb Space Telescope, the most powerful telescope in human history, we can peer through the time capsule spanning billions of years. Recently we had the most audacious insight into the farthest corner of the universe with the light only few hundred million years after the so called Big Bang being captured by James Webb Space Telescope.

It's in that spirit that one may expect that many of the astronomical and spatial voids will be filled in with Data Science and Machine Learning algorithms by leveraging an avalanche of data obtained by many sophisticated space telescopes, including James Webb Space Telescope, and other advanced tools.  In 2019, even before the James Webb Space Telescope was launched, space researchers were thrilled as they embraced the first glimpse of a doughnut-shaped super-massive Black Hole. It was a bright orange color Black Hole located circa 54 million lightyears away in a distant galaxy known as Galaxy Messier 87. The image of the massive Black Hole was collected by pre-James Webb-era Event Horizon Telescope, or EHT, Platform. EHT Platform is a constellation of telescopes that employ a special data collection methodology, called the Very Long Baseline Interferometry, to collect multitude of images at various resolution levels. Despite the powerful array of telescopes positioned around various regions of the world to get an optimal sneak-peak of Galaxy Messier 87 and its embedded supermassive Black Hole, the images obtained by EHT Platform were obscure and hazy, complicating the work of scientists in their quest to drive appropriate analysis and derive more precise conclusions about the supermassive Black Hole.

The data disjoints, insight incongruence and related pain points pave the way for a fertile ground for innovation through high-fidelity, high-scalability data simulation models for matters falling through the Black Hole. There came one of the most sophisticated Data Science and Machine Learning (DSML) modeling techniques in space research. Principal Component Interferometric Modeling, or PRIMO, technique runs tens of thousands of high-fidelity, high-scalable simulations that provide the requisite fillers for the voids in the images obtained from the EHT Platform. PRIMO simulations empowered the scientists to cross all the T’s and dot all the I’s, leading to eventual more precise construct of the Black Hole image. The Astrophysical Journal Letters has published research in its April 2023 edition.