Reference Match Types

Infosphere's Quality Product

Moderators: chulett, rschirm

Post Reply
srds2
Premium Member
Premium Member
Posts: 66
Joined: Tue Nov 29, 2011 6:56 pm

Reference Match Types

Post by srds2 »

Hi, I am trying to understand the different types of Reference Match grouping. I have read the documentation but couldn't understand the difference between below types of matching and also if we are trying to get One input record matched with multiple Reference Records then what kind of Reference match type we should go for and Viceversa.

Many-to-one
Many-to-one Multiple
Many-to-one Duplicate

Can anyone help me to better understand these reference match types.

Thanks a lot in advance!
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

A two-source match identifies records in one source that have similar attributes to records in a second source (for example when enriching a data source from a reference source).
In one-to-one matching, one record from the data source can be assigned to only one record from the reference source, and vice versa. Each record pertains to a single individual or event.
You use many-to-one matching to match a single data source to a reference source. A reference record can match to many records on the data source.
The three types are well explained in the IBM InfoSphere Information Center. This page will answer your specific question.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
srds2
Premium Member
Premium Member
Posts: 66
Joined: Tue Nov 29, 2011 6:56 pm

Post by srds2 »

Thanks Ray for the information.

for one to one matching : Lets say we have 1 source record similar to two reference records then the second reference record will fall under which of the output category (Match, Clerical, Reference Duplicates, Reference Residual?) Because in one to one mathing one source record should be matched to only one reference record. I coldnt get this infornation in the documentation.

I read the scenarios given in the documentation but still couldnt understand the exact difference or when should we use Many to one Vs Many to one Multiple Vs Many to One Duplicate. Can you please provide me some more information on these two types?

Thanks a lot in advance.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

The only possible answers are (a) "it depends" on the characteristics of the data and (b) if you have this knowledge a priori then you probably should be using a many-to-one match rather than one-to-one.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply