When I learned about Association Rule Mining, the example was similar to https://www.geeksforgeeks.org/frequent-item-set-in-data-set-association-rule-mining/ and Frequent Itemsets & Association Rules – Apriori Algorithm

But I was asked a question with duplicated items in transactions like:

transaction_id, items
1, A A B
2, A B C C

So what is the support count for A?
Should I only count the number of transactions containing A, which is 2? And ignore the duplicates?

why downvote? Is it because it is not reasonable to have a duplicated column in a transaction? but this is an exercise question from some unreasonable teacher…