Titanic Dataset — Exploratory Data Analysis (EDA)

This repository contains a complete, step-by-step Exploratory Data Analysis (EDA) of the Kaggle Titanic: Machine Learning from Disaster dataset.
The analysis was conducted in Python using Pandas, Matplotlib, and Seaborn within Google Colab.

---

📂 Contents

• titanic_eda.ipynb — Full Jupyter Notebook with EDA steps.
• titanic_clean.csv — Cleaned dataset after preprocessing.
• figures/ — Saved visualizations from the notebook.

---

🔍 Key Analysis Steps

1. Data Loading & Overview
2. Missing Data Analysis & Cleaning
3. Feature Engineering (Title, Family Size, IsAlone, etc.)
4. Univariate Analysis — Distributions & counts for key features.
5. Bivariate Analysis — Survival by Sex, Class, Embarked.
6. Heatmaps — Survival rate pivot tables & correlation matrix.
7. Insights & Executive Summary — Actionable findings and next steps.

---

📊 Example Visualizations

• Survival rate by gender, passenger class, and port of embarkation.
• Age and Fare distributions with outlier detection.
• Correlation heatmap for numeric and engineered features.

---

---

✨ Author

Masum Abbas — Data Science Student at Air University, passionate about data analysis, visualization, and turning raw data into actionable insights.