Power BI - How to use Fuzzy Matching

Video gives an overview of how to use Fuzzy Matching logic in Power Bi. Below are few options that can be manually selected.

Similarity Threshold: This number can be from 0 to 1. It signifies of how closely similar the text strings should match in order to show in the output table. By default is set up to 0.8 i.e. 80% similar

Ignore Case: It indicates whether text values should be compared using case sensitivity. The default behavior is case insensitive

Match by combining text parts – This option is used if there’s a space between text strings, it’ll combine both text strings into one.

Maximum number of matches: This option controls the maximum number of matching rows that will be returned for each input row. For example, if you only want to find one matching row for each input row, specify a value of 1. The default behavior is to return all matches.

Transformation Table : This option allows users to specify another query that holds a mapping table, so that some values can be auto-mapped as part of the matching logic. For example, defining a two-column table with a “From” and “To” text columns with values “Microsoft” and “MSFT” will make these two values be considered the same (similarity score of 1.00) by the matching logic.