Forum Discussion
Incorrect similarity score calculation for text string comparison
I'm comparing two text strings via a fuzzy join function to get the similarity score.
However, I'm getting incorrect results for the similarity score.
If I'm not mistaken in my search for solution Fuzzy Match calculation explained , it should be based on the jaccard similarity score. However the results are not corresponding:
Example 'INDEMANS CHR SRL' vs 'INDEMANS CHRISTIAN'
Power Query: result 0,37
Online tool: result 0,55
How is this possible?
Thank you in advance.
Update: I've added a step removing spaces in both columns, making the result for these specific strings better (0,89). I'm then returning the maximum score of the two comparisons. However, I suppose normally the 'IgnoreSpace' should have covered this, which clearly is not the case)