Projects

 Identifying Potential Habitats Beyond Earth: A Multilayered Statistical  Analysis of NASA’s Confirmed Exoplanets


Duration  :  July 2023 - Present

Description :

T.B.A

Skills and Techniques : 

Acknowledgments : 

T.B.A

A Novel Sector-Based Algorithm for an Optimized Star-Galaxy Classification.

Duration  :  July 2023 - December 2023

Description :

Here we propose a novel methodology of optimizing Star - Galaxy Classification using Sonal Digital Sky Survey Data Release 18. 

Skills and Techniques : 

Acknowledgments : 

T.B.A

AstroML: Machine Learning Applications in Astronomy and Astrophysics

Description :

AstroML: Machine Learning Applications in Astronomy and Astrophysics is an in-depth exploration of the rapidly evolving intersection of machine learning and astronomical research. As an undergraduate physics student with a passion for astronomy, I have created this mini-book to provide fellow enthusiasts with an accessible introduction to the advanced techniques used in modern astrophysics research. This study material covers a wide range of machine learning methodologies, with a strong emphasis on their practical applications in solving real-world astronomical problems.

Throughout the guide, readers will be introduced to various machine learning algorithms, such as regression, classification, clustering, and deep learning, and will learn how these techniques can be used to analyze massive astronomical datasets, identify celestial objects, and uncover the hidden patterns within the cosmos. Each concept is reinforced with Python code examples, allowing you to gain hands-on experience in applying these techniques to real data.

Whether you're a student, researcher, or simply an astronomy enthusiast, this resource aims to provide you with a solid foundation in machine learning for astronomy and astrophysics, empowering you to explore the universe in an entirely new way.

 Optimization of a One-Dimensional Hypertelescope for Direct Imaging in Astronomy 


Course Name :   Applied Optimization  ( DSE 311 )

Course Instructor :  Dr.  Sujit Pedda Baliyarasimhuni

Duration  : One Semester ( Spring 2023 )

Description :

This course project involves the study and recalculation of results from the paper "Optimization of a One-Dimensional Hypertelescope for Direct Imaging in Astronomy." The goal is to optimize the relative positions of output pupils and the modulus of beams through each pupil in a linear array of telescopes, enabling the instrument to directly image exoplanets. By applying nonlinear optimization techniques in interferometric optical astronomy, the project aims to improve the capabilities of this hypertelescope design.

Skills and Techniques : 

Throughout this project, I have developed a range of technical skills and techniques, including :

Acknowledgments : 

 I would like to express my gratitude to my course instructor, Dr. Sujit Pedda Baliyarasimhuni, for his invaluable guidance and support throughout this project. Additionally, I would like to acknowledge the authors of the paper,Paul Armand , Joël Benoist , Elsa Bousquet , Laurent Delage , Serge Olivier , François Reynaud , for their insightful research and contributions to the field of astronomy. Finally, I am grateful to my peers for their constructive feedback and collaboration during the course of this project.


Data Analysis and Classification on NASA Near Earth Objects Dataset.


Description :

In this project, I worked on the NASA Near Earth Objects (NEOs) dataset to perform data analysis and classification. The goal of this project was to identify the various characteristics of NEOs and develop a model to predict their classification based on these characteristics.

To begin, I downloaded the dataset from the NASA website and performed data cleaning and preprocessing using Python libraries like Pandas, NumPy, and Scikit-learn. This involved removing missing values, normalizing the data, and converting categorical variables into numerical form.Next, I performed exploratory data analysis to gain insights into the distribution of the data and identify any correlations between the variables. I used visualizations like histograms, scatter plots, and heatmaps to better understand the data.After this, I trained and tested several machine learning models, including Logistic Regression, KNN, and Random Forest. I used the Scikit-learn library to implement these models and evaluated their performance using metrics like accuracy, precision, recall, and F1-score.After comparing the results of the models, I selected the Random Forest classifier as the best-performing model with an accuracy score of 0.95. This model showed a high degree of accuracy in predicting the classification of NEOs based on their characteristics.Finally, I analyzed the feature importance to identify the most significant variables that contribute to the classification. This information can be used to improve our understanding of the characteristics of NEOs and their potential impact on Earth.

Overall, this project provided valuable insights into the characteristics and classification of NEOs and demonstrated the usefulness of data analysis and machine learning in this field. The results can be used to improve our understanding of NEOs and their potential impact on Earth.

I am proud of the work that I have accomplished in this project and look forward to applying my newfound knowledge and skills in future projects related to data analysis and classification.

Citizen Scientist : Exoplanet Watch


Description :

I participated in a Citizen Science project that involves analyzing telescope data and creating light curves using the exotic code presented by NASA. Through this project, I developed skills in data analysis, data visualization, and Python programming. I also learned how to work with astronomical data and gained a deeper understanding of the observational techniques used in astronomy.

Exploring Cosmological Parameters with MCMC and Runge-Kutta Methods


Course Name :  Numerical methods and Programming  ( PHY 312 )

Course Instructor :  Dr. Nirmal Ganguli

Duration  : One Semester ( Spring 2023 )

Description :

 T . B . A

Skills and Techniques : 

Throughout this project, I have developed a range of technical skills and techniques, including :

Acknowledgments : 

I would like to thank Dr. Nirmal Ganguli for his guidencen and support throughout this project and I am also i would like to acknowledge my project partner Shreejith for his contributions to this project. 

Dysonian approach to SETI using EDA and Machine Learning (DSE 315) 


Course Name :  Data Science in Practice (DSE 315)

Course Instructor :  Dr. Parthiban Srinivasan

Duration  : One Semester ( Fall 2022 )

Description :

In this project, I learned about web scraping, exploratory data analysis (EDA), and machine learning. Using a Dysonian approach to SETI, we searched for stars that could potentially host artificial megastructures. We first created a chart of modeled absolute magnitude vs. observed absolute magnitude of 1000 random stars from our dataset using EDA and rigorous modeling. We then identified a set of anomalously dim stars in green, and a set of bright stars in orange. From this, we constructed the Hertzsprung–Russell (H-R) diagram and found that the anomalously dim candidates, particularly those in the main sequence or below it, exhibit stellar clustering.

Skills and Techniques : 

Throughout this project, I have developed a range of technical skills and techniques, including :

Acknowledgments : 

I would like to thank Dr. Parthiban Srinivasan for his guidencen and support throughout this project and I am also grateful to the authors of the paper " Dysonian Approach to SETI : A Fruitful Middle Ground ?" for their valuable contributions to the field of Astronomy in machine learning.