[Link].
com
hr@[Link]
Data Science with Python Internship – Task 3
Task Description
In Task-2, you analyzed the Titanic dataset for survival by gender and class.
Now, let’s perform a Mini Exploratory Data Analysis (EDA) on the Titanic dataset with
more advanced visualizations.
You will:
• Handle missing values better (imputation).
• Use groupby to find deeper insights.
• Create multiple visualizations to tell a story.
What You’ll Build (Requirements)
1. Clean dataset:
o Fill missing Age with mean.
o Drop irrelevant columns (like Cabin).
2. Analysis Questions:
o Survival rate by Age Group.
o Survival rate by Embarkation Port.
o Survival rate by Family size (SibSp + Parch).
3. Visualizations:
o Age distribution (histogram).
o Heatmap of correlations.
o Survival by family size (bar plot).
Deliverable
A Jupyter Notebook with cleaned data, multiple insights, and professional visualizations.
Why This Task Is Important
• Teaches you EDA (Exploratory Data Analysis), a must for every data scientist.
• Improves your data cleaning skills.
• Helps you build storytelling with data.
Sample Starter Code
Free Video Courses / Tutorials
• Data Science Crash Course – FreeCodeCamp
• Pandas Crash Course – Corey Schafer
• Seaborn Visualization Crash Course
Documentation
• Pandas Docs
• Matplotlib Docs
• Seaborn Docs
🛠 Tools
• Google Colab (Free Jupyter Notebook)
• Kaggle Datasets
• Titanic Dataset