This repository contains a hands-on project on predicting survival on the Titanic using machine learning. The project was developed as part of a lecture on machine learning and model training for beginners.
The goal of this project is to predict whether a passenger would survive the Titanic disaster based on various parameters such as age, gender, class, and more. The dataset used for training the model is the popular Titanic dataset, available on Kaggle.
The following machine learning algorithms were used:
- Decision Tree Classifier
- XGBoost Classifier
- Random Forest Classifier
- Logistic Regression
The Random Forest Classifier achieved the best accuracy of 82.06%.
titanic_survival.ipynb: Jupyter Notebook containing data preprocessing, feature engineering, and model training code.