Options

Support Vector Clustering

PichaPicha Member Posts: 2 Contributor I
edited November 2018 in Help

y3mH9jFwKDmbla8MUZXkchoCVv1bOl1qd2frocIiARYOtjDPSaEttRggepKFuRkz794FVIMhZbXVoda8Rpew9B3dahbmBSkSWArATdKkhIfrkU.jpg

 

I use distance performance operators for support vector clustering but the result has error. Following this picture to add in E-mail. I want to know I connect ports Right? 

 

I use support vector clustering operator for cluster items in supermarket. There are 2882 SKUs. And there are 5 criterial for clustering. 

Tagged:

Answers

  • Options
    IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder

    Hi,

     

    Distance-based cluster performance only worked for centroid-based cluster models like those delivered by k-Means or k-Medoids.

     

    Hope this helps,

    Ingo

  • Options
    PichaPicha Member Posts: 2 Contributor I

    What performance operator I use for SVC? I try to use all performance operator the result has error. 

  • Options
    IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder

    Hi,

     

    I am afraid there is none which works for SVC.  Part of the reason is that not many people are actually using SVC - it is notoriously hard to tune... :smileysad:

     

    Best,

    Ingo

Sign In or Register to comment.