First of all, let me give you a bit of context. I am into the final year of my undergraduate studies and as part of the final year, I have the opportunity to do a research project. So this is merely a concept I cooked up inside my head and I wanted to hear what the fellow devs think before I pitch this to my supervisor. Any feedback is welcome.

Let's take StackOverflow as an example, because why not? Currently, any type of feedback you give, doesn't matter who you are and how much reputation you have, you can have a 10k reputation or you can have 10 reputations, it's all taken the same.
What I basically want to do is change this and make these feedback a user gives, weigh according to their expertise and the context of the question.
For Example, let's imagine there is a FORTRAN question,

User A - Very experienced in FORTRAN.
User B - Very experienced in Java.

In this scenario, User A's feedback will have more weight than User B's feedback.

This can be used in upvoting/downvoting questions or answers, ranking the order of answers displayed.

How to come up with the expertise?

For that, I'm thinking of profiling the users based on their StackOverflow stats and maybe use their GitHub stats as well.

I hope you understood what I'm trying to say. Look at this concept as a pure research project.

Any feedback, good or bad is welcome, or if there are any improvements to be made.