Data Science in Python: Data Preparation & EDA

Transforming raw data into meaningful insight. One dataset at a time.

Overview

This repository captures my journey of building a strong foundation in data science through project-based learning in Python.

Rather than approaching data science as a collection of tools, this work focuses on developing a structured way of thinking — how to move from ambiguous problems to clear, data-driven insights.

Working as a Jr. Data Scientist on a simulated business case (Maven Music), I analyze customer data to understand patterns in churn and user behavior. Each notebook reflects not just technical execution, but the reasoning and decision-making behind it.

What This Work Represents

This repository is not just coursework. It represents:

• Learning how to translate business problems into analytical questions
• Developing the ability to work with imperfect, real-world data
• Building intuition for what to explore, what to question, and what to ignore
• Practicing how to extract signal from noise through structured analysis

Core Focus Areas

• Understanding the Data Science Workflow
• Supervised vs Unsupervised Learning
• Project Scoping Techniques
• Data Collection from Multiple Sources
• Data Cleaning and Preprocessing using Pandas
• Exploratory Data Analysis (EDA)
  • Filtering and grouping data
  • Data visualization
  • Correlation
• Data Preparation for Machine Learning
  • Feature engineering
  • Dataset merging & transformation

Tools & Technologies

• Python
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Google Colab

Course Pathway

The work in this repository follows a structured progression:

1. Foundations of Data Science
2. Problem Scoping & Analytical Thinking
3. Data Gathering & Cleaning
4. Exploratory Data Analysis
5. Project-Based Application
6. Preparing Data for Machine Learning

Key Outcomes

Through this work, I have developed the ability to:

• Approach data problems with a clear analytical framework
• Clean, structure and validate messy datasets
• Perform EDA that leads to meaningful, actionable insights
• Prepare datasets for downstream machine learning tasks
• Communicate findings through structured, visual narratives

Who This Is For

• Aspiring Data Scientists building foundational skills
• Data Analysts transitioning into more advanced roles
• Python learners aiming to work with real-world datasets

ⓘ Note

This repository is continuously evolving.