[AccessD] A2003: Checking for Similar Names and or Addresses

[Jim Dettman]

Depends on what the goal is.   If it's only to check for exact matches, then
a collapse of the strings involved (removing all white space) and
concatenated against an index suffices.

If it's more of a "fuzzy" type of analysis that you want to use, then there
are various approaches, like soundex (i.e.  One name is "Jim Dettman" vs
"Jim Debtman").

Weighting would be throwing in comparing the different components, name vs
address.

On a straight compare though, I've always fully collapsed the strings, then
do a find on an index.

Jim.

-----Original Message-----
From: AccessD [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of
Darren
Sent: Wednesday, May 13, 2015 11:47 PM
To: 'Access Developers discussion and problem solving'
Subject: [AccessD] A2003: Checking for Similar Names and or Addresses

Hi Team
I appreciate this may trigger a right vs. wrong discussion. All comments
appreciated but this is for a Pro-Bono Project so there are no funds and my
time is not infinite.
So a perfect solution is not required. Just a working one :-)
So...
I have inherited a dB with approx 12K names and addresses for a community
project I am working on.
Data entry has been on home built Access dBs and Excel spreadsheets and data
veracity is poor.
I can easily identify nearly 1000 records with same first and last names so
there appear to be a lot of duplicate entries.
We will handle that and will sanitise the old data. However does anyone have
a quick and easy "test" where I can quickly check completed names (First and
Last) once entered, to see if we have exact or similar matches already in
the dB? 
I really would like the same thing with addresses?
I have Googled and I have seen some very complex weighting routines etc -
Too tricky for me and beyond the scope of this project (and my skill set).
I just need to present the users with a list of exact matches (I can do that
now) and some potentials, in a list that they can then decide if it's ok to
proceed or not.
Many thanks in advance.
Darren.
-- 
AccessD mailing list
AccessD at databaseadvisors.com
http://databaseadvisors.com/mailman/listinfo/accessd
Website: http://www.databaseadvisors.com