Hungarian Algorithm Matching in Power Query?

[Anonymous]

Hi - I am trying to use power query to create the most optimal matches in a dataset. Think of it as the college dorm assigment problem. I have survey responses from a group of people and I want to match individuals with the most opposite responses. The issue is rather than create a unique score for each person, the individual's responses are compared to every other person to create a score for every possible pairing. I have been able to create a matrix with the participants on the X and Y axis and their pairing score. Now I need to select the pairings with the highest scores, without any of the participants being selected more than once. 

After some digging, it appears to be a perfect use case for the hungarian aglorithm, does anyone know how to do this in Power Query, or maybe have the Python code to do it?       

Thanks in advance

[Anonymous]

@v-yingjl those are great links, thank you! 

[v-yingjl · Accepted solution]

 Hi @Anonymous ,

About Hungarian Algorithm in python, I think you could refer these two simliar articles which introduce it in details:

1. The basic principle of Hungarian algorithm and its implementation in Python 
2. Hungarian Algorithm 

Best Regards,
Community Support Team _ Yingjie Li
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.