Fuzzy Match 2 Columns with score

[Lauritsjust]

Hi All

I am using the solution provided in Comparing Records between two columns for similarity

However, the solution is not quite what I am looking for. The solution provided only looks up the pairs provided in Field1 and Field2, where i would like the solution to find the "highest" match from field 1 and field 2. 

Input:

Output:

Desired output: 

I would like the result to show me that (Field1:8) and (Field2:5) are almost the same, and the names can thus be corrected in my full dataset. 

I also need to filter for unique values.

Any ideas?