How to train model that compares similarities of phone contacts list?

Question

Let's say I have data of phone numbers:

(i.e. a collection that each of its indexes contains a list of phone numbers. (array of arrays))

What I want to is a method to recognize "similar contacts". i.e. how the data in the "i" index is similar to the data in the "j" index.

I want to answer this question using a trained ml model.

I guess one way of doing it is using "supervised learning" that uses a classifier that compares "vectors" that represent each contact list.

I'm new to ml and python also, so can you please advise how it can be done?

Answer 1

You don't need to use ML for this problem. You can use string distance which measures how similar two string are (obviously convert the phone numbers to strings before you use the algorithm). There is more info here and a pretty good python package to help here.