I am trying to figure what type of ML this falls into.
Let's say I have 10,000 users and 200 lists - Each user can belong to as many lists as they want. What type of ML system would evaluate the likelihood of a user belonging to list based on all the other users they share memberships with on all other lists.
(Hopefully that makes sense, I think this is a solved problem and I am just looking for a pointer to solution)
Sounds almost like a recommendation problem. Let's see if replacing "user" with "item" and "list" with "user" in your original statement helps:
Let's say I have 10,000 items and 200 users - Each item can belong to as many users as they want. What type of ML system would evaluate the likelihood of an item belonging to user based on all the other items they share memberships with on all other users.
In a recommender setting, you usually recommend an item to users by trying to predict the probability of a user consuming certain item. In your setting, you'll be trying to predict the probability of a list "consuming" certain user
My first try would be to set up 200 KNN models, since the memberships are not exclusive. A dense neural network with member properties as input and 200 probability outputs should work too.
