Kaggle's Spaceship Titanic Competition
The task in this competition is to predict whether a passenger was transported to an alternate dimension during the Spaceship Titanic’s collision with the spacetime anomaly. To help with these predictions, a set of personal records recovered from the ship’s damaged computer system is given.
In the code above I start with Data Cleaning and Exploratory Data Analysis. Then I build a preprocessing pipeline with the insights from those steps and fit the data with the following models: Support Vector Machine, Random Forest, Ada Boosting and Gradient Boosting. Lastly, I investigate which features are being used to make the predictions and reduce the input space to avoid overfitting. By using an ensemble of all models I was able to get an accuracy of 0.80126.
Visualizations were made with Plotly, data manipulation with Pandas and machine learning tasks with Scikit-Learn.