Real-world data is rarely ready to analyse. It arrives with missing values, duplicates, outliers, inconsistent formats, and messy text, and analysts spend a large share of their time fixing these problems before any real analysis begins.
This module shows how to clean data with Pandas: taking a first look at a dataset, standardizing values such as phone numbers, filtering out rows you don't need, and handling nulls and duplicates. You'll also practise wrangling: matching similar strings, and merging, grouping, and reshaping data from several tables into one analysis-ready dataset.
With clean data in hand, you'll be ready for the next module, Exploratory Data Analysis, where you start looking for patterns.
A step-by-step walkthrough of cleaning a customer list in Pandas: taking a first look at the data, cleaning and standardizing phone numbers, and filtering down rows of data.
A 5-hour project-based course that includes cleaning a Google Play Store dataset (null values, duplicates, and outliers), matching company names by string similarity, and merging and grouping data from several dataframes.