❗ How it works:

The clustering algorithm uses phonetics (soundex) and orthographic (Levenshtein-Damerau) resemblance to match names and surnames.

You can adjust the threshold slider to control the tolerance on orthographic errors (the higher the threshold, the smaller the tolerance).

To begin with, you must first load data. "Load sample data" will just give you a non-realistic example to play with.

Pre-computed datasets are available (select "Load from Server", then select the file you want and finally click "Load").

Finally, you can also chose to load your own JSON file.