Read this story carefully
I am a data science, machine learning, and deep learning enthusiast. But along with this, I am also a big enthusiast in astronomy and space. I have always seen that machine learning and deep learning have solved most of the modern days' problems be it financial or linguistic or medical and healthcare-related. A.I. has always surpassed our expectations.
But could A.I. algorithms find its way to the solution in case of astronomy?. Yes, it could. But the most challenging thing which I found, in this case, was the availability of sufficient data.
Although the struggle to find appropriate data to work with was not much longer. I finally came across this dataset which is officially maintained by Jet Propulsion Laboratory of California Institute of Technology which is an organization under NASA. With this, I also found my way to solve the most interesting problem, which is to predict asteroid diameter with A.I. algorithm.
I researched this topic and found that the prediction of asteroid diameter is not an easy job. There had been various methods to solve the estimation of asteroid diameter, every method trying to surpass its previous one. More detail about the latest methods is discussed in my research paper(link is below).
I tried solving this problem with various ML algorithms like XGboost regressor, Gradient Boost regressor, Ada-Boost regressor, Random Forest Regressor and finally concluded that Multi-Layered Perceptron regressor could solve this problem with the least error possible and higher accuracy.
The time when I found this dataset, I was simply a B.Tech third-year student. I practiced data-science and machine learning problems at Kaggle, and with that, I also had a very keen interest in publishing research papers. I came across various research papers on ML, DL, and AI and always wondered one day I could also publish my very own research paper. Finding my work to be totally unique on the Internet, I decided to publish a paper on this, and I finally published it.
Link to JPL database - https://ssd.jpl.nasa.gov/sbdb_query.cgi
Link to my paper - https://goo.gl/qiunPq
"Prediction of Asteroid Diameter with the help of Multi-layer Perceptron Regressor".2019.International
Conference on Computer Science, Industrial Electronics(ICCSIE-2019) organized by Industrial Electronics and
Electrical Engineers Forum
Your Mission
The time when I wrote this paper, I did not knew much about feature engineering, Deep Learning, different neural network architecture and Other machine learning algorithms like LighGBM, catboost( In simple words almost a noob😖😖). I did not even knew how to write a decent research paper. I just had a rush to have my very own research paper. Your job is to surpass my work and make a better model with higher accuracy and the least error possible than mine . Not only to solve the prediction of asteroid diameter, but this dataset has the information to solve various other problems related to astronomy. The link to the JPL database which I have mentioned above would also help you to extract data about the comets with updated information about them. So, you could also have your research on comets.
Link to JPL database - https://ssd.jpl.nasa.gov/sbdb_query.cgi
Last but not the least, all my best wishes for your research on Astronomy.
Acknowledgements
I wouldn't be here without the help of NASA. I heartfully thank NASA and JPL for maintaining such an wonderful database.
Using data-science and ML to solve an astronomical problem wouldn't have been possible because of their hard work.
Inspiration
I have always been Inspired by astronomers and space scientist and I hope my work and contribution would also inspire you to work on astronomical problems. Not only this but there are various astronomical problems in the outer space which await A.I. to solve them.