I am currently looking at how social groups are connected. There are group, id, and order variables. My goal is to find the groups that overlap (easy enough with "duplicates tag id, gen(dup)" and combine them all to a new group (this is the hard part) . I have looked at this post that seems similar, but doesn't help too much.
I could do this by hand, but I have millions of data points and at least 100000 groups. Some of these groups get very big as well.
How can I identify and tag these unique networks? A network might not just be the intersection of two groups, but three intersections among four groups, etc.
This probably doesn't seem clear, but I am having difficulty articulating.
Below is a photo showing what I want (kind of).
I could do this by hand, but I have millions of data points and at least 100000 groups. Some of these groups get very big as well.
How can I identify and tag these unique networks? A network might not just be the intersection of two groups, but three intersections among four groups, etc.
This probably doesn't seem clear, but I am having difficulty articulating.
Below is a photo showing what I want (kind of).
Comment