Function Repository Resource:

PartitionByCooccurence

Find the best-fitting partitions for a list of groups using co-occurrence counts

Contributed by: Ryan Bell

ResourceFunction["PartitionByCooccurence"][subgroups]

partitions the subgroups list based on the co-occurrence frequency of list items.

Details and Options

This function is used to partition a collection of items into subgroups based on the co-occurrence frequency of the list items.

Examples

Basic Examples (1)

For all the elements in a set of lists, find a set of disjoint subsets containing all the elements:

In[1]:=

Out[1]=

Applications (1)

Self-assembling communities (1)

Suppose you have a list of user IDs which are conversing on different threads. On one thread, there’s a discussion between users {"Alice","Bob","Carol","Dan"}, on another thread between {"Alice","Bob"} and on a third thread, users {"Carol","Dan"} are chatting. Using this information, let’s find the best way to organize all participants into groups based on these self-assembling communities: