Which node should be used to evaluate the results of making a cross validation with a method of clustering as k-means? If the data do not include a class.
thanks
Knime 12345.
If you want to evaluate k means cross validation is not the way. You are right clustering don’t have classes. If you want to evaluate your clustering you can return to initial node and split it, clustering first split, and named the results. Then do the same with the other split. After that, use a crosstab node and do a chi-square test. If you can reject null hipothesis then you can conclude that clustering was well done, because most of subject belong to the same clustering in both parts.
Could you send a workflow or a picture with a sample?, With your instructions I do not know how to create the workflow.
Another cuestion there is a node for the "sum of square error" for a clustering algorithm
Thank you very much in advance
Hi, can I ask some help. with my undergrad thesis? I have these electric load data for the month of April. And I am using Knime's Kmeans node, to cluster the data to determine its base, mid, and peak load. How can I check the validity of my cluster? An image of my workflow is attached.