Book Image

Python Data Cleaning Cookbook

By : Michael Walker
Book Image

Python Data Cleaning Cookbook

By: Michael Walker

Overview of this book

Getting clean data to reveal insights is essential, as directly jumping into data analysis without proper data cleaning may lead to incorrect results. This book shows you tools and techniques that you can apply to clean and handle data with Python. You'll begin by getting familiar with the shape of data by using practices that can be deployed routinely with most data sources. Then, the book teaches you how to manipulate data to get it into a useful form. You'll also learn how to filter and summarize data to gain insights and better understand what makes sense and what does not, along with discovering how to operate on data to address the issues you've identified. Moving on, you'll perform key tasks, such as handling missing values, validating errors, removing duplicate data, monitoring high volumes of data, and handling outliers and invalid dates. Next, you'll cover recipes on using supervised learning and Naive Bayes analysis to identify unexpected values and classification errors, and generate visualizations for exploratory data analysis (EDA) to visualize unexpected values. Finally, you'll build functions and classes that you can reuse without modification when you have new data. By the end of this Python book, you'll be equipped with all the key skills that you need to clean data and diagnose problems within it.
Table of Contents (12 chapters)

Combining DataFrames vertically

There are times when we need to append rows from one data table to another. This will almost always be rows from data tables with similar structures, along with the same columns and data types. For example, we might get a new CSV file containing hospital patient outcomes each month and need to add that to our existing data. Alternatively, we might end up working at a school district central office and receive data from many different schools. We might want to combine this data before conducting analyses.

Even when the data structure across months and across schools (in these examples) is theoretically the same, it may not be in practice. Business practices can change from one period to another. This can be intentional or happen inadvertently due to staff turnover or some external factor. One institution or department might implement practices somewhat differently than another, and some data values might be different for some institutions or missing...