BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
ycenycute
Obsidian | Level 7

I have several questions regarding association node. 

 

1. Does it require at least one ID variable and one Target variable? Why is that? What does ID and Target do here?

2. What does maximum items mean? I can find the descriptions in the document, but still don't understand. Can someone explain? Like what is a four-way association? Why when there is sequence variable, it is reset to 2? 

  • Maximum Items — Use the Maximum Items property to specify the maximum size of any given item set that you want to consider in an association. For the example, the default value of 4 items indicates that up to four-way associations are performed. The minimum number of Maximum Items is 1. Permissible values are nonnegative integers. If you set Maximum Items to 1 and your data source does not contain a sequence variable, then Maximum Items is reset to 2.
  •  
1 ACCEPTED SOLUTION

Accepted Solutions
Amod1
SAS Employee
Ref: https://documentation.sas.com/doc/en/emref/14.3/n16x97j506upgin1l90wrfc1rg0l.htm

1. I think of ID variable as a customer, Target variable as Item he purchases and sequence variable as his visit number, or the date of purchase.
2. An item set contains items purchased. A frequent item set is that item set which occurs (purchased) frequently by the ID (customer). Now, maximum items tells us about the size of the item set. If 4, an item set can contain at the most 4 items.
3. 4-way associations: possible ways in which we can combine them to form any association rule. We can take 1 or 2 or 3 or 4 (all) of them in creating an association.
4. Now, in case when you have sequence (date/time) variable, you can associate 1 item (target) from 1st visit of same customer (ID) with any other item from the subsequent visits (purchase dates). That’s why we need at least 1 item in maximum items parameter. However, if we don’t have sequence variable, we are restricted to find the associations within the same visit (purchase date), as there is only one visit by that customer. In that case, association requires at least 2 items, one on LHS and the other on RHS. That’s why If you set Maximum Items to 1 and your data source does not contain a sequence variable, then Maximum Items is reset to 2.

View solution in original post

1 REPLY 1
Amod1
SAS Employee
Ref: https://documentation.sas.com/doc/en/emref/14.3/n16x97j506upgin1l90wrfc1rg0l.htm

1. I think of ID variable as a customer, Target variable as Item he purchases and sequence variable as his visit number, or the date of purchase.
2. An item set contains items purchased. A frequent item set is that item set which occurs (purchased) frequently by the ID (customer). Now, maximum items tells us about the size of the item set. If 4, an item set can contain at the most 4 items.
3. 4-way associations: possible ways in which we can combine them to form any association rule. We can take 1 or 2 or 3 or 4 (all) of them in creating an association.
4. Now, in case when you have sequence (date/time) variable, you can associate 1 item (target) from 1st visit of same customer (ID) with any other item from the subsequent visits (purchase dates). That’s why we need at least 1 item in maximum items parameter. However, if we don’t have sequence variable, we are restricted to find the associations within the same visit (purchase date), as there is only one visit by that customer. In that case, association requires at least 2 items, one on LHS and the other on RHS. That’s why If you set Maximum Items to 1 and your data source does not contain a sequence variable, then Maximum Items is reset to 2.

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!

Submit your idea!

How to choose a machine learning algorithm

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.

Discussion stats
  • 1 reply
  • 450 views
  • 1 like
  • 2 in conversation