February 5, 2009 at 3:11 pm
I'm trying to validate the results I've gotten with an Association Rules Data Mining Model. A given itemset (Product A, Product B) has a Support of 723. I thought to myself, let me run an sql query that counts the customers who have both Product A and Product B. The count I get from that query is 1023. So I decided to use the Drill-Through on the Mining Model Viewer. That also gives me the results "Query execution completed with 1023 rows fetched".
I thought Support was the frequency count? The Microsoft Association Algorithm Technical Reference http://msdn.microsoft.com/en-us/library/cc280428.aspx states, "Support, which is sometimes referred to as frequency, means the number of cases that contain the targeted item or combination of items. Only items that have at least the specified amount of support can be included in the model." Therefore, I would expect that the support would be 1023, not 723. Can anyone explain to me why there is a difference? I'm going to have to explain it to the end-user.
Thanks,
Cindy
February 6, 2009 at 6:33 am
Anyone?
February 6, 2009 at 3:01 pm
Would duplicates affect your query? In other words does "support" include duplicate items?
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply