I'm doing and MBA and using Association Node for that. I'm playing around with the Support Percentage property. Base on SAS Help, "The support percentage figure that you specify refers to the proportion of the largest single item frequency, and not the end support."
My understanding is this. I did a product ranking, counting how many transactions for each product divided by total transaction. Example:
Product 1 = 60%
Product 2 = 10%
Product 3 = 2%
Product 4 = 1%
If I set support percentage of 2%, then Products 1-3 will be considered for rule-building. But I'm getting product 4 in the final rules, as left hand. Anyone has idea why this product comes out? Or is my understanding not correct?
Thanks a lot!
Below option determines the minimum level of support for a rule as a percentage of the number of baskets in the input data set.
PCTSUP"|"SUPPCT"|"SUP_PCT"|"PCTSUPPORT
In your case rules alone or combination of Product 3 & Product 4 will be chosen if their combination also suffice this percentage.
I guess Rules with combination of Product 3 & Product 4 may not have required SUPPCT so dropped and rules consisting of Product 4 seems to have required SUPPCT so final chosen.
Hope I answered your query.
Thanks and Regards,
Kiran Bhole
I am just taking rules with two items so simply 1->2 or 2->3 or 3-> or 4->1. So I am not expecting any 4->1. Rule combination of 3&4->1 is not of interest and not considered.
What the help documentation says about the Support Percent property is, and i quote, "not the end support". So the results go me a bit confused.
Which documentation are you referring?
Online Help documentation of SAS Enterprise Miner: https://go.documentation.sas.com/?docsetId=emref&docsetTarget=n16x97j506upgin1l90wrfc1rg0l.htm&docse...
Under "Association Node Train Properties: Association" it says the following:
Support Percentage — When the Support Type property is set to Percentage, use the Support Percentage property to specify the minimum level of support to claim that items are associated (that is, occur together in the database). Permissible values are real numbers between 0 and 100. The support percentage figure that you specify refers to the proportion of the largest single item frequency, and not the end support. The default frequency is 5%.
Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
Register now!
Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.