Wiki

As defined above, Record-Based matching is depended on all the attributes in each individual record of the previously defined identity. This means that the first name, last name, and school code attribute for each record in the source must have an exactly matching record in the identity it is being compared against.

This is how the matching takes place:

1. Record "1|Sam|Jacobs|05071982|SC11" is selected.

2. The first name, last name and school code are noted:

  a.    Sam, Jacobs, SC11

3. Record

#!xml

"A^source1.3|B^Sam|C^Jacobs|D^05071983|E^SC21"

is selected from the identity.

4. The first name, last name, and school code are noted:

  a.    Sam, Jacobs, SC21

5. The attributes from the source and the identity record are compared

  a.    Does Sam = Sam?

      i.    Yes

  b.    Does Jacobs = Jacobs?

      i.    Yes

  c.    Does SC11 = SC21?

      i.    No

6. Since all 3 attributes do not match this record from the identity is discarded and the next record from the identity is selected:

a.

#!xml

A^source1.4|B^Sam|C^Jacobs|D^05071982|E^SC11"/>

7. The attributes from the source and the identity record are compared

  a.    Does Sam = Sam?

      i.    Yes

  b.    Does Jacobs = Jacobs?

      i.    Yes

  c.    Does SC11 = SC11?

      i.    Yes

8. Since all the matching rules are satisfied, it is determined that the source record is a match and it is now part of the identity.

9. The new identity would look like this:

#!xml

<Identity Identifier="VUKPECCU0KHBJUDO" CDate="2011-06-14">
    <References>
       <Reference Value="A^source1.3|B^Sam|C^Jacobs|D^05071983|E^SC21"/>
       <Reference Value="A^source1.4|B^Sam|C^Jacobs|D^05071982|E^SC11"/>
       <Reference Value="A^source1.5|B^Samual|C^Jacobs|D^05071982|E^SC11"/>
         <Reference Value="A^source2.1|B^Sam|C^Jacobs|D^05071982|E^SC11"/>
    </References>
</Identity>

10. Once the first record is matched, the process aborts for this identity and selects the next source record:

  a.    2|Samual|Jacobs|05071982|SC21

11. If you apply the same process to this source record you will see that the records do not match so a new identity will be created. The final identities will look like this:

#!xml

<Identity Identifier="VUKPECCU0KHBJUDO" CDate="2011-06-14">
    <References>
       <Reference Value="A^source1.3|B^Sam|C^Jacobs|D^05071983|E^SC21"/>
       <Reference Value="A^source1.4|B^Sam|C^Jacobs|D^05071982|E^SC11"/>
       <Reference Value="A^source1.5|B^Samual|C^Jacobs|D^05071982|E^SC11"/>
         <Reference Value="A^source2.1|B^Sam|C^Jacobs|D^05071982|E^SC11"/>
    </References>
</Identity>

#!xml

<Identity Identifier="SUGPQC4UBKPBAU9O" CDate="2011-06-14">
    <References>
       <Reference Value="A^source2.2|B^Samual|C^Jacobs|D^05071982|E^SC21"/>
    </References>
</Identity>

Please note that the Record-Based matching will produce fewer consolidations in general but requires less processing and memory which speeds up the overall processing time.

Back to OYSTER Reference Guide page

Click Prev Attribute Based page

Click Next 4-Reference Sources page