Lets say I got this df.
df={'name': [Peter, Alf, Bob, Birgit, Julia], 'categories': [amazon, paypal, Walmart, [], Walmart], 'usage': [bill number 12345 Simon, bill number 23451 Smith, bill number 75647 Simon, bill number 52547 Jill, bill number 95847 Jon]}
How can I cluster the categories, so that the program shows me e.g amazon is online store, PayPal a online payments system and so on. I know this df is very small but can anyone help me in the beginning. I started to look after NER, but don't know if it is the right choice. The final result of my task is a program that categorize new data automatically. Thanks in advance. ps: I can't use spacy in my environment.