🧭 Titanic Dataset — Exploratory Data Analysis (EDA)

📘 Overview

This project performs an in-depth Exploratory Data Analysis (EDA) on the famous Titanic dataset to uncover the patterns and social dynamics that influenced passenger survival.
Using Python, Pandas, and Seaborn, we visualize, summarize, and interpret the data to reveal how age, gender, ticket class, and fare shaped survival outcomes aboard the RMS Titanic.

---

🎯 Objectives

1. Understand the structure and content of the dataset.
2. Identify missing values, outliers, and data inconsistencies.
3. Explore distributions of key variables using histograms and boxplots.
4. Analyze survival patterns across gender, passenger class, and embarkation points.
5. Discover correlations and relationships using heatmaps and pairplots.
6. Summarize key insights and actionable findings for future modeling.

---

🧩 Tools & Libraries Used

• Python
• Pandas – Data handling and preprocessing
• NumPy – Numerical operations
• Matplotlib & Seaborn – Data visualization
• ReportLab – Generating PDF reports

---

📊 Key Steps Performed

1️⃣ Data Understanding

• Used .info(), .describe(), and .value_counts() to inspect structure and missing data.
• Found missing values in Age, Cabin, and Embarked columns.

2️⃣ Univariate Analysis

• Plotted histograms and boxplots for Age and Fare.
• Observed Fare’s heavy right-skew (few extremely high fares).
• Most passengers were male, 3rd class, and embarked from Southampton.

3️⃣ Bivariate Analysis

• Compared Survival with gender, class, and embarkation point.
• Females had the highest survival rate (~74%), while males had the lowest (~19%).
• 1st class passengers survived more than 3rd class.

4️⃣ Multivariate Analysis

• Used sns.pairplot() and sns.heatmap() to visualize correlations.
• Strong negative correlation between Pclass and Survival (−0.34).
• Positive correlation between Fare and Survival (+0.26).

5️⃣ Observations

• Wealth and gender heavily influenced survival.
• Younger and higher-class passengers were prioritized during rescue.
• Large families had lower odds of survival compared to moderate family sizes.

---

💡 Insights Summary

Factor	Impact on Survival
Gender	Females had significantly higher survival chances.
Class	1st class passengers survived most; 3rd class the least.
Fare	Higher fares correlated with higher survival rates.
Age	Younger passengers showed slightly higher survival odds.
Embarked	Passengers from Cherbourg (C) had higher survival.

---

🧠 Key Learnings

• Real-world data reflects human and social patterns, not just numbers.
• Data visualization simplifies complex insights into clear, actionable stories.
• EDA helps identify key variables for predictive modeling and feature engineering.

---

📄 Project Deliverables

File	Description
Task-5 Exploratory Data Analysis.ipynb	Jupyter notebook containing code, plots, and step-by-step explanations.
task_5_Finding_Summary.pdf	A professionally formatted, humanized EDA report.
README.md	Project overview and documentation (this file).

---

🚀 How to Run

1. Clone the repository

   git clone https://github.com/rushofcode/Exploratory-Data-Analysis-EDA.git
   cd Exploratory-Data-Analysis-EDA.git
   

2. Install dependencies

bash Copy code pip install pandas numpy matplotlib seaborn reportlab

3.Run the notebook bash Copy code jupyter notebook Task-5 Exploratory Data Analysis.ipynb View the final PDF report in your repo directory.

👩‍💻 Author Rushikesh Palekar