When we de-dupe a file, what we do is create a key of LastName*FirstName[1,3]*zipcode or Country (If a lot of the firstnames are similar, we will extend the [1,3] to be the entire first name (which usually includes a Middle initial also).
Anyway...We are purchasing a canned software package which does some fuzzy logic to du-dupe (This was the easiest, and fastest method to improve our matching) But just curious.....I wanted to try to do a % matched on all fields (name, address, city, state, zip , cntry) eliminating St, Ave, Suite, etc, so only the unique pieces of the data remain. Only problem is I couldn't figure a way to make a key that I could compare to, in order to see if this name exists, or what percentage. Anyway, how are others processing de-dupes? George Gallen Senior Programmer/Analyst Accounting/Data Division [EMAIL PROTECTED] ph:856.848.9005 Ext 220 > The Wyanoke Group http://www.wyanokegroup.com ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
