Winning space race with data science
Himanshu Narwal
Data Scientist
Data Visualizer
ML Engineer
Python
scikit-learn
SQL
Project description
Develop and evaluate machine learning algorithms to determine the likelihood of reusing the initial stage of a rocket in a space launch.
Background and context
SpaceX advertises Falcon 9 rocket launches on its website with a cost of 62 million dollars.
Other providers cost upwards of 165 million dollars each.
Much of the savings is because SpaceX can reuse the first stage.
Therefore, if we can determine if the first stage will land, we can determine the cost of a launch.
This information can be used if an alternate company wants to bid against SpaceX for a rocket launch.
The problem
Use public information to predict if SpaceX will reuse the first stage.
Methodology
Data collection methodology
SpaceX REST API.
Web scraping related Wikipedia pages.
Perform data wrangling
Cleaning, filtering and dealing with missing values.
Applying one-hot encoding to categorical features.
Perform EDA using visualisation and SQL
Perform interactive visual analytics using Folium and Plotly Dash
Perform predictive analysis using classification models
Developing logistic regression, k-nearest neighbors, support vector machines and decision tree models.
Comparing them to determine which model performs the best.
Results
Launch success yearly trend.
Total successful launches by site.
Correlation between payload and outcome.
Classification accuracy.
