Pandas Textual Data Manipulation

Introduction

In this lab, we will explore how to manipulate textual data using Python's Pandas library. You will learn how to convert string characters to lowercase, extract parts of strings, replace string values and more using various built-in Pandas methods.

VM Tips

After the VM startup is done, click the top left corner to switch to the Notebook tab to access Jupyter Notebook for practice.

Sometimes, you may need to wait a few seconds for Jupyter Notebook to finish loading. The validation of operations cannot be automated because of limitations in Jupyter Notebook.

If you face issues during learning, feel free to ask Labby. Provide feedback after the session, and we will promptly resolve the problem for you.

Import Necessary Libraries and Data

Let's start by importing the Pandas library and loading the data we will use for this tutorial.

## Import necessary libraries
import pandas as pd

## Load the data
titanic = pd.read_csv("data/titanic.csv")

Convert String Characters to Lowercase

Next, we will convert all characters in the Name column to lowercase. We'll use the str.lower() method to achieve this.

## Convert all characters in the 'Name' column to lowercase
titanic["Name"] = titanic["Name"].str.lower()

Extract Surnames from Full Names

Now, let's create a new column Surname that contains the surname of the passengers. We'll achieve this by extracting the part before the comma in the Name column.

## Split the 'Name' column on comma and extract the first part
titanic["Surname"] = titanic["Name"].str.split(",").str.get(0)

Extract Specific Passenger Data

Next, let's extract the passenger data for the countesses on board of the Titanic. We'll use the str.contains() method to find rows where the Name column contains the word 'Countess'.

## Find rows where 'Name' contains 'Countess'
countesses = titanic[titanic["Name"].str.contains("Countess")]

Find the Longest Name

Let's find out which passenger of the Titanic has the longest name. We'll use the str.len() method to get the length of each name, and the idxmax() method to find the index of the longest name.

## Get the length of each name
name_lengths = titanic["Name"].str.len()

## Find the index of the longest name
longest_name_index = name_lengths.idxmax()

## Get the longest name
longest_name = titanic.loc[longest_name_index, "Name"]

Replace Values in a Column

Finally, let's replace the values in the Sex column: 'male' with 'M' and 'female' with 'F'. We'll use the replace() method for this.

## Replace 'male' with 'M' and 'female' with 'F' in the 'Sex' column
titanic["Sex_short"] = titanic["Sex"].replace({"male": "M", "female": "F"})

Summary

In this lab, we have seen how to manipulate textual data using Python's Pandas library. We have learnt how to convert string characters to lowercase, extract parts of strings, find specific rows based on string contents, find the longest string, and replace string values. This knowledge is very useful in data preprocessing, a crucial step in data analysis and machine learning.

Pandas Textual Data