-
Objective: Leveraged machine learning techniques to improve exoplanet detection accuracy, enhancing model performance from 87% to 93%.
-
Dataset: Analyzed a dataset of 5,087 stars, each with 3,197 flux values recorded over time, which enabled the creation of light curves to detect potential orbiting exoplanets.
-
Tech Stack:
-
Libraries: Utilized Pandas, Seaborn, Numpy, and Matplotlib for data manipulation, visualization, and statistical analysis.
-
Machine Learning Model: Implemented K-Nearest Neighbors (KNN) with a 99.02% validation accuracy to identify the nearest stars, enabling reliable exoplanet detection.
-
After KNN and handling the data embalance
-
