Options
"Visualise attributes within clusters"
Contributor II
I'm new to data mining and would mind some input on how to display clusters and their 'contents'.
I have a dataset
test1 test2 test3 test4 testX
true false frue false false
true true etc...
I have created clusters using expectation maximisation from this and was wondering how I visualise what is in the clusters (i.e. cluster 1 has test 1,3,4; cluster 2 has test 3,4) etc.
Should I have organised my data differently? What is my next step? I'm using RapidMiner 5.
I have a dataset
test1 test2 test3 test4 testX
true false frue false false
true true etc...
I have created clusters using expectation maximisation from this and was wondering how I visualise what is in the clusters (i.e. cluster 1 has test 1,3,4; cluster 2 has test 3,4) etc.
Should I have organised my data differently? What is my next step? I'm using RapidMiner 5.
Tagged:
0
Answers
usually clusters are grouping the rows of a data set. Within RapidMiner these are called Examples. The columns of the data set are used as Attributes of these examples. These attributes are used during clustering to find the most similar examples and to group them in one cluster.
So I guess, you might need to rotate your data. If it's the case, you might use the Transpose operator to do that.
Greetings,
Sebastian