Welcome! This is my attempt at the infamous Kaggle Titanic Machine Learning Challenge.
The goal of this project was to create a model that accurately predict the survival of Titanic passengers given certain data (aka features) regarding each passenger (Sex, Passenger Class, Cabin, etc). I found meaningful insights/relationships between certain features through data visualization. More specifically, I found basic relationships and slowly complicated the complexitiy of my analysis through data visualization. I cleaned/manipulated the data to properly train the model, and then use that trained model on unseen data. In other words, given unseen data of new passengers, the trained model attempts to predict survival rate of those new passenegers.
For now, my best submission is 76.55% accurate, meaning that I built a model that correctly predicted the survival of 76.55% of Titanic passengers. I'm looking to improve that score through feature engineering, cross-validation, etc. Stay tuned! :)