Titanic AI 🚢

An artificial intelligence model to predict passenger survival on the Titanic disaster using machine learning.

Overview

This project implements a complete machine learning pipeline to analyze the Titanic passenger dataset and predict survival outcomes. The model achieves 84% accuracy using Random Forest classification with comprehensive data preprocessing and feature engineering.

Dataset

The project uses the famous Titanic dataset containing:

• PassengerId: Unique identifier for each passenger
• Survived: Survival outcome (0 = No, 1 = Yes) - Target variable
• Pclass: Passenger class (1st, 2nd, 3rd)
• Sex: Gender of passenger
• Age: Age in years
• SibSp: Number of siblings/spouses aboard
• Parch: Number of parents/children aboard
• Fare: Passenger fare
• Embarked: Port of embarkation (C = Cherbourg, Q = Queenstown, S = Southampton)

Machine Learning Pipeline

1. Data Preprocessing

• Missing Value Imputation:
  • Age: filled with median value (28.0 years)
  • Embarked: filled with most frequent port (Southampton)
• Feature Removal:
  • Cabin: dropped due to 77% missing values
  • Name & Ticket: removed as non-predictive features
• Categorical Encoding:
  • Sex: binary encoding (male=0, female=1)
  • Embarked: one-hot encoding with drop_first=True

2. Model Training

• Algorithm: Random Forest Classifier
• Parameters: 100 estimators, random_state=42
• Train/Test Split: 80/20 ratio
• Cross-validation: Built-in bootstrap sampling

3. Model Performance

• Accuracy: 84% on test set
• Robustness: Consistent performance across different runs
• Interpretability: Feature importance analysis available

Requirements

pip install pandas scikit-learn

Dependencies:

• Python 3.13+
• pandas 2.3.0+
• scikit-learn 1.7.0+

Usage

# Clone the repository
git clone https://github.com/SpaceBuddy231/Titanic-AI.git
cd Titanic-AI

# Run the model
python main.py

Project Structure

Titanic-AI/
├── main.py          # Complete ML pipeline
├── data/
│   ├── train.csv    # Training dataset (891 passengers)
│   ├── test.csv     # Test dataset (418 passengers)
│   └── gender_submission.csv # Sample submission format
└── README.md        # Project documentation

Key Features

✅ Complete Data Preprocessing Pipeline
✅ Missing Value Handling
✅ Feature Engineering & Encoding
✅ Machine Learning Model Training
✅ Model Evaluation & Metrics
✅ Clean, Documented Code

Results

The model successfully processes the raw Titanic dataset and produces:

• Clean numerical features suitable for ML algorithms
• Zero missing values in the final dataset
• 84% prediction accuracy on survival outcomes
• Robust performance with Random Forest classification

Future Enhancements

Potential improvements for higher accuracy:

• Advanced feature engineering (family size, titles from names)
• Hyperparameter tuning with GridSearch
• Ensemble methods (Voting, Stacking)
• Cross-validation optimization