October 1, 2014 at 7:46 am
Hello
We have multiple databases with tables with client's data, name/dob/ssn.
We are looking for app which will match all clients in all dbs/tables and build list of ids for each client in each table.
i.e. table1 have following clients:
1 / Alex / 1-1-1960 / 5665
2 / John / 1-2-1960 / 5465
table2 have following clients:
1 / John / 1-2-1960 / 5465
2 / Max / 1-3-1960 / 5454
So result of work of app should be table or csv file with following data:
Name / DOB / SSN / Table1 ID / Table2 ID
Alex / 1-1-1960 / 5665 / 1 / NULL
John / 1-2-1960 / 5465 / 2 / 1
Max / 1-3-1960 / 5454 / NULL / 2
Again, we are looking for 3rd party app which can be easy customize to match different dbs / tables / fields.
Anybody know that type of app?
Much thanks,
Alex.
October 1, 2014 at 8:46 am
you are overthinking this if you think you need an app for that.
no "app" is needed for this, as the data entry for any app that tried to do this
is the same effort required for deciding the mapping criteria for as the query
SELECT T1.ID AS Table1ID,
T1.Name AS Table1Name,
T1.DOB AS Table1DOB,
T1.SSN AS Table1SSN,
T2.ID AS Table2ID,
--,T2.Name As Table2Name
--,T2.DOB AS Table2DOB
T2.SSN AS Table2SSN
FROM DatabaseOne.dbo.Table1 T1
INNER JOIN DatabaseTwo.dbo.OtherTable T2
ON T1.Name = T2.Name
AND T1.DOB = T2.DOB
AND T1.SSN = T2.SSN
Lowell
October 1, 2014 at 11:22 am
Lowell (10/1/2014)
you are overthinking this if you think you need an app for that.no "app" is needed for this, as the data entry for any app that tried to do this
is the same effort required for deciding the mapping criteria for as the query
Thanks for SQL query, but it will not work, since name can be spelled with errors, or in different ways, i.e. Alex, Aleks, Alexandr, Alexander.
For Alex vs Aleks we can use SOUNDEX, but not for Alex vs Alexander.
October 1, 2014 at 2:08 pm
onixsoft (10/1/2014)
Lowell (10/1/2014)
you are overthinking this if you think you need an app for that.no "app" is needed for this, as the data entry for any app that tried to do this
is the same effort required for deciding the mapping criteria for as the query
Thanks for SQL query, but it will not work, since name can be spelled with errors, or in different ways, i.e. Alex, Aleks, Alexandr, Alexander.
For Alex vs Aleks we can use SOUNDEX, but not for Alex vs Alexander.
don't join on names. try SSN and date of birth, or just SSN, and ignore names.
if you have nay otehr columns that uniquely idnetify a person(email, drivers license number, etc) use that.
don't bother trying to join on names.
Lowell
October 2, 2014 at 6:35 am
don't join on names. try SSN and date of birth, or just SSN, and ignore names.
if you have nay otehr columns that uniquely idnetify a person(email, drivers license number, etc) use that.
don't bother trying to join on names.
We can't avoid names, source databases are mess, some clients have only names, some have names and dob, ssn.
We need app like Redgate data comparer, which will show which records can be matched and allow us to make final decision are they the same clients or not. So if clients have same name , but very close DOBs, it can be the same client but with wrong DOBs, so app should display results and we will determine if they are the same client and which one have correct dob.
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply