Idea about a function for error correction when importing Names

[fmow]

Here's an idea for a function in FileMaker: when you're importing names that might have slight spelling differences with the already existing ones, one might want to create a function that would compare an imported name and surname with all the available, correct ones. Then it would enter the closest match from the database at a cut off: i.e. 80% to 100% similar. 

 

Does that ring a bell to anyone?

 

Do you guys think that would be a worthwhile function?

[Matthew F]

Sure, the option for "fuzzy" matching would be nice, particularly if you could specify the degree of matching necessary.  Are you thinking of creating a custom function, or are you hoping that Filemaker will add this as a new feature? 

[jbante]

Look into Soundex and Hamming distance, and see if either of those might get close to what you're looking for.

[fmow]

Sure, the option for "fuzzy" matching would be nice, particularly if you could specify the degree of matching necessary.  Are you thinking of creating a custom function, or are you hoping that Filemaker will add this as a new feature? 

 

I hadn't though of it as a feature to add, but it sounds great if they could. Yeap, I was thinking of going about it on my own, with a function and I 've been told to look at the Levenshtein algorithm 

Look into Soundex and Hamming distance, and see if either of those might get close to what you're looking for.

Great thanks! I ll also be looking at Damerau-Levenshtein distance algo.