Need some help in the approach to text mining and then market basket analysis


#1

I have a dataset on tickets that are manually classified and stored in a column say ticket_pattern from the description field. I need to automate the process. There is another column which is the ticket type.

My understanding is first do the text mining on the description field group by the ticket type and get the most frequent words then do the apriori analysis with the ticket_pattern after getting the frequent words there as well?

or at each row, do the text mining for both columns and perform the apriori analysis?