Welcome to the repository for our Machine Learning-based Exoplanet Detection System, developed to classify celestial bodies using real-world astronomical data. This project harnesses supervised learning algorithms to distinguish between confirmed exoplanets and non-planetary candidates based on features from the NASA Exoplanet Archive.
Check out the interactive web application powered by Streamlit: 👉 exoplanetsml.streamlit.app
This project aims to automate the detection of exoplanets using publicly available datasets. It employs classification algorithms to analyze planetary and stellar characteristics such as:
- Orbital period
- Planetary radius
- Stellar radius
- Effective temperature
After preprocessing and training multiple models, the K-Nearest Neighbors (KNN) algorithm was selected for its interpretability and strong performance on numerical data.
- Classify whether a given object is a confirmed exoplanet.
- Utilize real data from NASA for training and evaluation.
- Build an intuitive front-end interface for public engagement using Streamlit.
- Cleaned and preprocessed exoplanet dataset
- Machine Learning model (KNN) for classification
- Data visualization using Plotly
- Web deployment with Streamlit for live interaction
- Lottie animations for enhanced user experience
The dataset is obtained from the NASA Exoplanet Archive, a science-driven, peer-reviewed repository containing thousands of confirmed and candidate exoplanet entries. The dataset includes features relevant for machine learning such as:
pl_orbper(orbital period)pl_rade(planet radius in Earth radii)st_rad(stellar radius)st_teff(stellar effective temperature)pl_letter(planet name/classification)
- Algorithm: K-Nearest Neighbors (KNN)
- Libraries:
pandas,numpy,sklearn,plotly,streamlit
Welcome to the repository for our Machine Learning-based Exoplanet Detection System, developed to classify celestial bodies using real-world astronomical data. This project harnesses supervised learning algorithms to distinguish between confirmed exoplanets and non-planetary candidates based on features from the NASA Exoplanet Archive.
Check out the interactive web application powered by Streamlit: 👉 exoplanetsml.streamlit.app
This project aims to automate the detection of exoplanets using publicly available datasets. It employs classification algorithms to analyze planetary and stellar characteristics such as:
- Orbital period
- Planetary radius
- Stellar radius
- Effective temperature
After preprocessing and training multiple models, the K-Nearest Neighbors (KNN) algorithm was selected for its interpretability and strong performance on numerical data.
- Classify whether a given object is a confirmed exoplanet.
- Utilize real data from NASA for training and evaluation.
- Build an intuitive front-end interface for public engagement using Streamlit.
- Cleaned and preprocessed exoplanet dataset
- Machine Learning model (KNN) for classification
- Data visualization using Plotly
- Web deployment with Streamlit for live interaction
- Lottie animations for enhanced user experience
The dataset is obtained from the NASA Exoplanet Archive, a science-driven, peer-reviewed repository containing thousands of confirmed and candidate exoplanet entries. The dataset includes features relevant for machine learning such as:
pl_orbper(orbital period)pl_rade(planet radius in Earth radii)st_rad(stellar radius)st_teff(stellar effective temperature)pl_letter(planet name/classification)
- Algorithm: K-Nearest Neighbors (KNN)
- Libraries:
pandas,numpy,sklearn,plotly,streamlit
├── mlcollege.py # Main Streamlit app code ├── requirements.txt # Dependencies for deployment ├── Animation.json # Lottie animation used in the app └── README.md # This file
Welcome to the repository for our Machine Learning-based Exoplanet Detection System, developed to classify celestial bodies using real-world astronomical data. This project harnesses supervised learning algorithms to distinguish between confirmed exoplanets and non-planetary candidates based on features from the NASA Exoplanet Archive.
Check out the interactive web application powered by Streamlit: 👉 exoplanetsml.streamlit.app
This project aims to automate the detection of exoplanets using publicly available datasets. It employs classification algorithms to analyze planetary and stellar characteristics such as:
- Orbital period
- Planetary radius
- Stellar radius
- Effective temperature
After preprocessing and training multiple models, the K-Nearest Neighbors (KNN) algorithm was selected for its interpretability and strong performance on numerical data.
- Classify whether a given object is a confirmed exoplanet.
- Utilize real data from NASA for training and evaluation.
- Build an intuitive front-end interface for public engagement using Streamlit.
- Cleaned and preprocessed exoplanet dataset
- Machine Learning model (KNN) for classification
- Data visualization using Plotly
- Web deployment with Streamlit for live interaction
- Lottie animations for enhanced user experience
The dataset is obtained from the NASA Exoplanet Archive, a science-driven, peer-reviewed repository containing thousands of confirmed and candidate exoplanet entries. The dataset includes features relevant for machine learning such as:
pl_orbper(orbital period)pl_rade(planet radius in Earth radii)st_rad(stellar radius)st_teff(stellar effective temperature)pl_letter(planet name/classification)
- Algorithm: K-Nearest Neighbors (KNN)
- Libraries:
pandas,numpy,sklearn,plotly,streamlit