Replace values in Pandas dataframe using regex

While working with large sets of data, it often contains text data and in many cases, those texts are not pretty at all. The text is often in very messier form and we need to clean those data before we can do anything meaningful with that text data. Mostly the text corpus is so large that we cannot manually list out all the texts that we want to replace. So in those cases, we use regular expressions to deal with such data having some pattern in it.

We have already discussed in the previous article how to replace some known string values in dataframe. In this post, we will use regular expressions to replace strings that have some pattern to it.

Using Dataframe.replace() Function

Problem #1: You are given a dataframe that contains the details about various events in different cities. For those cities which start with the keyword ‘New’ or ‘new’, change it to ‘New_’.

Solution: We are going to use regular expression to detect such names and then we will use Dataframe.replace() function to replace those names.

Output :

                City        Event   Cost
2018-02     New York        Music  10000
2018-04      Parague       Poetry   5000
2018-06    New Delhi      Theatre  15000
2018-10       Venice       Comedy   2000
2018-12  new Orleans  Tech_Summit  12000

Now we will write the regular expression to match the string and then we will use Dataframe.replace() function to replace those names.

Output :

                 City        Event   Cost
2018-02     New_ York        Music  10000
2018-04       Parague       Poetry   5000
2018-06    New_ Delhi      Theatre  15000
2018-10        Venice       Comedy   2000
2018-12  New_ Orleans  Tech_Summit  12000

As we can see in the output, the old strings have been replaced with the new ones successfully.  

Problem #2: You are given a dataframe containing details about various events in different cities. The names of certain cities contain some additional details enclosed in a bracket. Search for such names and remove the additional details.

Solutioncontaining: For this task, we will write our own customized function using regular expression to identify and update the names of those cities. Additionally, We will use Dataframe.apply() function to apply our customized function on each values the column.

Output :

                      City        Event   Cost
2018-02    New York (City)        Music  10000
2018-04            Parague       Poetry   5000
2018-06  New Delhi (Delhi)      Theatre  15000
2018-10             Venice       Comedy   2000
2018-12        new Orleans  Tech_Summit  12000

Now we will write our own customized function to match the description in the names of the cities.

Output :

                City        Event   Cost
2018-02    New York         Music  10000
2018-04      Parague       Poetry   5000
2018-06   New Delhi       Theatre  15000
2018-10       Venice       Comedy   2000
2018-12  new Orleans  Tech_Summit  12000