Identifying (similar) Names and Addresses


I have 50,000 names and addresses from multiple sources.

The wil be at least 30% duplication.

I.e. One specific name and address may be there more than once but many NOT be 100% identical.

John Smith, 1 High Street, London
John Smith, 1 The High Street London

Can anyone guide me to a utility which would identify names/addresses that are not quite 100% matched.

Any thoughts out there?
Patrick O'DeaAsked:
Who is Participating?
Scott McDaniel (Microsoft Access MVP - EE MVE )Connect With a Mentor Infotrakker SoftwareCommented:
That's going to be difficult to do.

You could use something like a Soundex algorithm to "rank" each one compared to the others. Essentially this would give you the greatest chance of duplicates for each entry, and you could then decide what to do with them.

Here's the wikipedia take on the Soundex stuff:

Essentially it involves replacing the characters in a string with numeric values, and then comparing the results. There are many different types of these algorithms, for various purposes. One example is this:

Consider the word "Cranston"

You keep the first letter ("C"), and then remove all other vowels, and any occurrence of letters y, h and w, so you're left with this:


You then assign values to the next 3 items. Using the wikipedia method, that would be:


The letter "r" is = 6, the letter "n" is = 5 and the letter "s" = 2.

You'd do the same for all the strings (and you could go out further than 3 letters if you'd prefer), and store this value in a column in that table. You then sort by that column, and you can see immediately which strings are most closely related.

Allen browne has one here: It uses a setup very much like what is described in the wikipedia link.
Patrick O'DeaAuthor Commented:
Thanks, I will experiment.  (I heard about soundex years ago).
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.