Data Cleaning

# Description

Data cleaning project with pandas on a customer dataset with intentionally dirty data. Learn to detect and fix the most common real-world problems: null values, duplicate rows, inconsistent text (spaces, capitalization), incorrect data types, outliers in ages and balances, invalid emails, and mixed date formats. 80% of time in ML is spent cleaning data — this project builds the foundation.

$ ./deploy.sh $ git clone

# Key features

$ Detection and visualization of null values with heatmaps

$ Removal of duplicate rows

$ Text cleaning: spaces, capitalization, inconsistent formats

$ Data type conversion with pd.to_numeric and pd.to_datetime

$ Outlier handling with IQR method and business rules

$ Null value imputation with median and mode

# Gallery

Desktop view

Mobile view

# Technologies used

Python pandas matplotlib seaborn