I have the music record dataset that I want run apriori on it. The dataset set is described here Apriori Algorithm using Python: music records | Kaggle
The workflow I created is simple. I grouped by user and choose aggregation method as artist (unique concatenate). So at the end I have a row for each user with the artist list that each user listened to (user’s music history). I excluded sex and country. Then I used create collection column node to convert the artist list to collection and used apriori node to generate association rules. Unfortunately I’m getting an empty set of rules although I configured the min support to as low as 0.1. Please help! What’s wrong with my approach!
Hello @barayh ,
Welcome to the KNIME Community!
Could you please share your workflow so that we can have a look at it?
Thanks in advance. This is my initial WF…
Week 9 - Association Rules.knwf (19.7 KB)
This is 10% and it is unlikely a rule will be able to cover that. In your linked example they go a low as 1% - 0.01 that is.
I did go to lower than that too but unfortunately I didn’t get any rules!
hi @barayh ,
simply made some changes on your workflow to the groupby node and setting of the confidence and support values learner.
Week 9 - Association Rules-MM.knwf (84.6 KB)
OHHH THANK YOU SO MUCH! It is working finally
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.