Pandas convert name where ID is the same and the name is different?

Question

I am trying to clean up name values where I have the following situation.

     ID  name
1     1    Company
2     1    Company, LLC

I would like to normalize it so I have only one name like so:

     ID  name
1     1    Company
2     1    Company

Answer 1

This will keep the first element of each group and broadcast it along the entire size of your dataframe:

df
Out[22]: 
   ID         name
0   1      Company
1   1  Company,LLC
2   2   Companybbb
3   2  Company,LLC
4   3   Companyccc
5   3  Company,LLC

df.groupby('ID')['name'].transform('first')
Out[21]: 
0       Company
1       Company
2    Companybbb
3    Companybbb
4    Companyccc
5    Companyccc
Name: name, dtype: object

Answer 2

For your example:

df.loc[df.name == 'Company, LLC', 'name'] = 'Company'

You can use this same method repeatedly to remap a sequence of values. As mentioned by MattR , FuzzyWuzzy can help you find strings that are likely candidates for being identical, if you want to identify more potential matches.