Titanic Dataset Analysis – Exploratory Data Analysis (EDA)

A complete data exploration project using Python, Pandas, Seaborn, and Matplotlib to analyze the Titanic dataset.

---

Overview

This project is part of my Data Analytics Internship at Elevvo Pathways.
The goal was to perform Exploratory Data Analysis (EDA) on the classic Titanic dataset, uncover survival patterns, calculate key metrics, and visualize insights professionally.

---

Workflow

1️⃣ Data Cleaning

• Handled missing values:
  • Age → filled with median
  • Cabin → dropped due to high missing rate
  • Embarked → filled with mode
• Corrected data types and ensured consistency across features.

2️⃣ Exploratory Data Analysis (EDA)

• Calculated key metrics (KPIs):
  • Overall Survival Rate
  • Survival Rate by Gender
  • Survival Rate by Passenger Class
  • Survival Rate by Embarkation Port
  • Average Age by Survival
• Group-based insights to understand patterns in survival.

3️⃣ Visualization

• Created professional charts to visualize key insights:
  • Barplots for survival by gender and passenger class
  • Horizontal Barplot for clearer class comparison
  • Stacked Histogram with KDE for age distribution by survival
  • Clustered Barplot (Catplot) for survival by class & gender
  • Correlation Heatmap for numeric feature relationships

Key Learnings

• Transform raw data into actionable insights
• Communicate patterns and relationships effectively through KPIs and visualization
• Practice Python, Pandas, Seaborn, Matplotlib in a professional data analytics workflow

---

Tools & Libraries

• Python
• Pandas
• NumPy
• Seaborn
• Matplotlib

---

Dataset

The dataset used is the Titanic: Machine Learning from Disaster from Kaggle:
🔗 https://www.kaggle.com/c/titanic

---

Author

[Sayed Esmail] – Data Analytics Intern at Elevvo Pathways