Exploring the Titanic Disaster with Machine Learning

This project applies machine learning techniques to the Titanic Dataset to predict whether a passenger survived or not.

Project Workflow

Data preprocessing and cleaning

Exploratory data analysis

Feature engineering

Model training and comparison

Model evaluation

Data Preprocessing

Removed irrelevant features: PassengerId, Name, Cabin, Ticket

Filled missing values (Age with median, Embarked with mode)

Encoded categorical variables (Sex with label encoding, Embarked with one-hot encoding)

Created a new feature: FamilySize = SibSp + Parch

Handled outliers using the IQR method

Applied feature scaling for some models

Models Used

Logistic Regression, KNN, Decision Tree, Extra Trees, Random Forest, Gradient Boosting, XGBoost, LightGBM, and Support Vector Machine.

Best Model

Support Vector Machine achieved the best performance with about 87% accuracy.

Evaluation

Model performance was evaluated using accuracy, precision, recall, F1-score, and a confusion matrix