It might be that it’s the end of my day but I have a problem I just cannot seem to figure out! I wish to create a matrix showing the overlap of different attributes as below (apologies if I counted any up wrong)

So far, I unpivoted my data to get the header rows with each attribute for each person. However, I am really struggling then matching it up with the row names to create the perfect square matrix. The eventual goal is to use it as source data to create a chord diagram in Tableau which I understand needs a square table like the one above.

Unfortunately your desired output as shown above could be “counted any up wrong”. Which makes it little bit of a guess if this wf matrix_of_different_attributes.knwf (42.4 KB) is what your are looking for. Because based on your input, my output is a little bit different. But take a look, and see if it helps you along.

I had a thought on the way home from work that I have probably asked the wrong question as well. Sorry! In my dataset, there are 29 different attributes and each record could have any number of them or none at all. So, the question is probably a lot more complex than the one I have posed! The ultimate goal is to create a visualisation of how the different attributes overlap, like having an enormously complex Venn diagram (I’ve tried an UpSet plot; it’s far too unwieldy to read)

It might be worth leaving this one be for a while whilst I think about it…however, I will definitely investigate your cross-joiner workflow example because I think that will be very useful for other tasks Thank you!