Machine Learning Project: Titanic Survival Prediction Using XGBoost

This project focuses on predicting passenger survival on the Titanic dataset using Machine Learning techniques.

The work includes:

- Advanced data preprocessing and feature engineering

(Title extraction, family size, fare per person, cabin encoding)

- Handling missing values and outliers

- Exploratory data analysis (EDA)

- Training a classification model using XGBoost

- Hyperparameter tuning and model optimization

Model Performance:

- Achieved accuracy of approximately 87%

Key Features Engineered:

- Title grouping from names

- FamilySize and IsAlone

- Fare normalization and clipping

- Age imputation using group-based strategies

This project demonstrates my ability to combine data analysis, feature engineering, and machine learning to build accurate predictive models.

Skills used:

- Python (Pandas, NumPy)

- Data Cleaning & Feature Engineering

- Machine Learning (XGBoost)

- Model Evaluation

Available for machine learning, data analysis, and predictive modeling projects.