The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

what is the difference between a FolderView and Centroid View?

charansaireddy2charansaireddy2 Member Posts: 1 Learner I
edited December 2018 in Help

I have perfromed a Cluster Analysis using the K-means operator. I have a result. But I am not able to differentiate between the Centroid view and Folder View. Does both of them show the same reusult or are they completely different? I am posting pictures of my result.Screenshot (142).pngcentroid viewScreenshot (143).pngFoler View

Tagged:

Answers

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

    Hi @charansaireddy2 - that's a good question. The "centroid view" shows you the value of the centroid for each cluster for each attribute. For example, in the Iris tutorial data set, you get this centroid table:

     

    Screen Shot 2018-05-21 at 10.15.08 AM.png

     

    Now if you make a simple scatterplot of the data, the centroids become clear:

     

    centroid.pngIris data set with three clusters and centroids indicated by X's.'

    (note that cluster_0 is blue/iris-virginica, cluster_1 is red/iris-setosa, cluster_2 is green/iris-versicolor).

     

    The "folder view" is simply a list of the id values of each element in the cluster:

     

    Screen Shot 2018-05-21 at 10.19.31 AM.pngIris data set clusters - note that id_50 is in cluster_1, id_51 is in cluster_0, id_52 is in cluster_2, and id_53 is in cluster_0 againScreen Shot 2018-05-21 at 10.20.01 AM.pngFolder view - note that id_51 and id_53 are both in cluster_0's folder

     

    Scott

     

     

     

     

     

Sign In or Register to comment.