i want to apply cluster on a data set and then apply feature selection on each cluster

imparveenimparveen MemberPosts:1Learner I
edited December 2018 inHelp

I am working on health Data set.What i am trying to do is that i want to make 2 cluster of this data set and then on each cluster i want to apply different feature selection methods .Using Rapid miner how can I use each cluster so that i can seperately apply feature selection techniques on both the clusters.

Tagged:

Answers

  • sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM ModeratorPosts:2,959Community Manager

    hello@imparveenwelcome to the community. Without looking at your XML I can just speak in generalization about this. Basically you would run a clustering algorithm on your data set (e.g. k-means) first. The output of this will be a new attribute called "cluster":

    Screen Shot 2018-09-24 at 9.51.03 AM.png

    If you then want to work on each cluster separately, I would just use Filter Examples:

    Screen Shot 2018-09-24 at 9.53.20 AM.pngScreen Shot 2018-09-24 at 9.53.00 AM.png

    XML of that process is here:










    <参数键=“repository_entry”值= " / /样品s/data/Iris"/>


















    output with only cluster_1
    output with only cluster_2



    Scott

Sign InorRegisterto comment.