Data Cleaning & EDA
Master the pandas toolkit for cleaning messy real-world data and conducting exploratory data analysis. Handle missing values, fix data types, remove duplicates, and find patterns in any dataset.
Why learn Data Cleaning & EDA?
Real datasets are never clean, and most of the actual work in data analysis happens before any chart gets built. Data Cleaning & EDA teaches the pandas toolkit for fixing missing values, duplicates, wrong types, and messy strings, then moves into exploratory data analysis: finding patterns and asking the right questions before you build anything.
It is 2 modules and 10 lessons, fully hands-on: data cleaning with pandas, covering missing values, duplicate detection, fixing data types, string cleaning, and outlier handling, and exploratory data analysis, covering the EDA mindset, a first look at any dataset, finding patterns with groupby, correlations and relationships, and a full EDA-in-practice lesson.
By the end you can take a messy CSV nobody has looked at yet and turn it into a dataset you actually understand.
Who this course is for
- Python for AI graduates who want to go deeper on the data-cleaning side specifically
- Aspiring data analysts who need pandas fluency for messy, real-world data
- Anyone who has opened a CSV full of missing values and blank cells and did not know where to start
- Developers preparing for the Data Analyst Portfolio track
What you'll build and practice
- Missing-value handling that does not silently distort your results
- Duplicate detection and removal on real, messy datasets
- Type fixes and string cleaning for columns that started out as a mess
- A repeatable process for taking a first look at any unfamiliar dataset
- Pattern-finding with groupby and correlation analysis, practiced end to end
What you'll learn
Data Cleaning & EDA is organized into 2 focused modules. By the end you'll be comfortable with:
- Data Cleaning with pandas
- Exploratory Data Analysis
Course curriculum
10 lessons across 2 modules. Lessons marked Free preview are readable without an account.
Module 1. Data Cleaning with pandas
Fix missing values, duplicates, type errors, and messy strings in real datasets
- 10mMissing ValuesFree preview
Detect, count, and handle null values — the first thing you check in any new dataset.
- 10mDuplicate DetectionFree preview
Find and remove duplicate rows — a critical step before any aggregation or count analysis.
- 10mFixing Data Types
Convert columns to the right dtype — the invisible fix that makes sorting, filtering, and math work correctly.
- 12mString Cleaning
Strip whitespace, fix case, split compound fields, and use regex for pattern-based cleaning.
- 12mOutlier Handling
Detect outliers with the IQR rule and decide whether to remove, cap, or keep them.
Module 2. Exploratory Data Analysis
Find patterns, summarise datasets, and ask the right questions before you build anything
- 10mThe EDA Mindset
Learn the questions to ask before writing any code — and why EDA is a conversation, not a checklist.
- 10mFirst Look at Any Dataset
The standard opening moves: .info(), .describe(), .value_counts() — understanding a dataset in under 5 minutes.
- 12mFinding Patterns with groupby
Slice data by categories to find where metrics differ — the core operation of exploratory analysis.
- 12mCorrelations and Relationships
Use df.corr() and heatmaps to find which variables move together — and which are independent.
- 15mEDA in Practice
A realistic end-to-end EDA walkthrough on a messy e-commerce dataset — asking and answering analytical questions.
Frequently asked questions
- Do I need to already know pandas before starting this course?
- Basic pandas familiarity, such as from Python for AI's Pandas Fundamentals module, helps you move faster, but this course re-teaches the operations you need in the context of cleaning, so it does not assume deep prior experience.
- Is this course only about cleaning data, or does it also cover analysis?
- Both, split evenly across its two modules. Module 1 is data cleaning specifically. Module 2 is exploratory data analysis: finding patterns, summarizing datasets, and asking good questions before drawing conclusions.
- How is this different from the data-cleaning lessons in Python for AI?
- Python for AI includes a data-cleaning module as one of seven inside a broader NumPy-to-machine-learning course. This track is a focused, 10-lesson deep dive purely on cleaning and exploratory analysis, with more practice on messier, more realistic scenarios.
- Does the course use real messy datasets, or already-clean examples?
- The lessons work with datasets that have the problems real data actually has: missing values, inconsistent types, duplicate rows, and messy string formatting, because those are exactly the problems the course teaches you to fix.
- What should I take after finishing Data Cleaning & EDA?
- Data Visualization & Business Metrics and Applied Statistics both build directly on a clean, well-explored dataset, and the Data Analyst Portfolio track is where all three come together in real projects.
Ready to start Data Cleaning & EDA?
Create a free account to track progress, earn XP, and get instant help from the AI tutor as you code.
Create your free account