We’re smarter together. Learn from this collection of community knowledge and add your expertise.

Association Discovery by Example in SAS® Enterprise Miner™

by SAS Employee Funda_SAS on ‎12-02-2015 11:21 AM (1,282 Views)


Download the Files (GitHub)


This tip, part of the Learn by Example using SAS® Enterprise Miner™  series, focuses on association analysis. SAS Enterprise Miner offers the following comprehensive resources to support association discovery:


  • Association node
  • Market Basket node
  • Link Analysis node
  • Path Analysis node


Each of these nodes has different strengths. For example, the Association and Path Analysis nodes enable you to perform sequence discovery by taking into account the time of the events. The Market Basket node can read taxonomy data to generate rules at multiple levels, and it is often blazingly fast compared to other nodes. While all of these nodes can read transactional data, the Link Analysis node can read row training data as well. Moreover, the Link Analysis node provides segmentation in addition to recommendations.


In order to understand which of these nodes would be a better fit for your recommendation systems problem, check out SAS Enterprise Miner Secure documentation. The secure site requires a user ID and password, which you can obtain by contacting SAS Technical Support directly. In order to expedite your request, please include “SAS Enterprise Miner documentation PDF” in the subject field of the form.


This tip focuses on the Association node. The flow diagram below shows the analysis of a transactional data to identify creditable associations between customer actions, such as having a baby and buying a new car,or buying a new home and opening a loan.



For a full description of the problem and how to analyze the Market Sequences data using SAS Enterprise Miner, see the pdf file in our GitHub repository. By using the XML file, you can also import the above diagram to your SAS Enterprise Miner project and play with it. 


Your turn
Sign In!

Want to write an article? Sign in with your profile.

Looking for the Ask the Expert series? Find it in its new home: communities.sas.com/askexpert.