Are the following operators same things: CreateAssociationRules & RuleInduction?

ameer2014ameer2014 Member Posts: 5 Contributor I
edited December 2018 in Help

Actually i am working on Apriori algorithm and want to measure the performance of it on my data. But ApplyModel operator doesn't accept rules, instead it requires model. That's why i had to use RuleInduction instead of CreatAssociationRules. But i want to make sure if these two operators are same.

Answers

  • earmijoearmijo Member Posts: 270 Unicorn

    Rule Induction is a classification model. 

    If I understand correctly your problem, you want to use the operator Apply Association Rules instead of Apply Model. 

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    Correct, as @earmijo states, the Rule Induction operator is a supervised learning operator and it requires a label.  It is based on the "Ripper" algorithm (see the operator help for more info).  Create Association Rules is an unsupervised learner that is looking for any associations among any attributes in your set, and formulating them in terms of support and confidence (once again, see the operator help for more details).  While there are some similarities, these two operators are most definitely not doing the same thing.

     

     

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
Sign In or Register to comment.