attribute autocorrelation not know if it is a right pace to ask but since i am working with knime to get the job done i asked it here:

i have a dataset which contains of rating data about 1600 movies(movielens dataset) and user demographic data..

i want to implement k-emans but the value ok k should be somehow determined based on attribute correlation meaning:


verify whether users with similar demographic profiles such as age, gender, occupation, location, etc. are grouped in the same clusters. We use the value of attribute autocorrelation as a parameter for determining the optimal number of clusters.
i just do not get the idea ....

can anyone plase help?

Hello m.a.najimi,

I just googled but all I could find was one paper that uses attribute autocorrelation, so could you please elaborate what this is?

There are many approaches on how to determine the number of clusters (Even a Wikipedia-article about it), maybe you can find a different approach? Or have a look at the Linear Correlation node?