Titanic - EDA & Data Preprocessing

This project explores the Titanic dataset from Kaggle to understand factors influencing passenger survival, and prepares the data for further analysis through preprocessing.

What this project covers

Exploratory Data Analysis (EDA)

• Dataset overview and structure check
• Missing value analysis
• Statistical summary
• Survival rate visualization
• Survival rate by gender and passenger class
• Age distribution analysis
• Correlation heatmap between numeric features

Data Preprocessing

• Handling missing values (Age, Embarked, dropping Cabin)
• Encoding categorical variables (Sex, Embarked)
• Outlier detection in Fare using the IQR method
• Feature scaling using StandardScaler

Dataset

Source: Titanic - Machine Learning from Disaster (Kaggle)

Tools used

• Python
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-learn