In a run-time environment, once a set of candidate records has been retrieved by the application program, the user’s application calls the
ssan3_match
function to score and further qualify the candidate records.
As an example of using this function, the following steps demonstrate how to match the Search Data for an "Ann Jackson Smith" living at "555 Harriman Plaza", with the File Data for an "Anne Jackson" living at "55 Harrison Plaza", using the "Resident" Match Purpose, a "Typical" Match Level and the "Tagged" method of passing data.
Under the
Mandatory Control
heading, click
Resident
in the list of Purposes. This adds the
Purpose=Resident
parameter to the
Controls
textbox. This will instruct SSA-NAME3 to use the Resident Match Purpose when comparing the search and file records entered below.
Under the
Optional Controls
heading, click
Typical
. This adds the
Match_Level=Typical
parameter to the
Controls
textbox. The text in the
Controls
will be,
PURPOSE=Resident MATCH_LEVEL=Typical
.
Select
Person_Name
from the Field Type dropdown box. This will add the text
*Person_Name*
to the Search Data text box.
Type the name
Ann Jackson Smith
immediately following the last asterisk.
Select
Address_Part1
from the Field Type dropdown box. This adds the text
*Address_Part1*
directly after the name just typed in the Search Data box.
Type the street address
555 Harriman Plaza
immediately following the last asterisk.
Select
End of Data
from the Field Type dropdown, or type three asterisks (***) after the address just entered.
Click
Tab
over to the File Data box.
Follow steps 3 through 7 above to create a file record with the Name Anne Jackson and the street address 55 Harrison Plaza.
Click the
Call
button. The results will appear in the
Decision
and
Score
boxes under the File Data as illustrated below:
The score will be a value between 0 and 100. The Match Decision is based on the pre-set score thresholds for this Purpose and Match Level. A Decision of "R" means the records would be Rejected, "U" means "Undecided" and "A" means Accepted.
If you select the
Hex
checkbox, the following screen appears with the hexadecimal equivalent of the string:
button causes the Match call to be invoked for the same search and file data, for each of the available match levels. This shows a comparison of the Decisions and Scores for the different Match Levels.
Click the
Match Level Comparison
button after entering the search and file data above will display the results below:
You can refine the Accept and Reject score limits by entering + or - values in the Accept Limit and/or the Reject Limit text boxes (in the Optional Controls section of the left handpane). For more information, see the Controls section in the
API REFERENCE guide
.
If you select the
Hex
checkbox, the following screen appears with the hexadecimal equivalent of the string:
Click the
Back to Match
button to return to the normal Match screen.
Combining Purposes
Match Purposes can be combined using the operators AND, OR and NOT. Match Filters can also be used in combination with Match Purposes. For information on how and why to use filters and combine purposes, refer to the Advanced Controls section in the
API REFERENCE GUIDE
.
The point and click functionality of the Workbench environment currently only supports simple combination of purposes (one operator with no match level overrides). To set up more complex multi-purpose controls, the details must be manually entered into the Controls text box.
For example, to set up the Controls for the following simple multi-purpose definition using point and click items in the left-hand-pane:
PURPOSE=(Individual OR Resident)
Click
Individual
in the left-hand pane, then click the operator
OR
, then click
Resident
. The following screenshot shows the construction of the Controls:
Match Explain
The internal logic of the match process can be viewed using the
Explain
button. This will open a dialog which presents the logic of the last match call performed in the form of a tree.
The score returned by match is the sum of the scores of the individual fields. As can be seen in the example above, 2 fields participate in the match process,
Address_Part1
and
Person_Name
.
Address_Part1
contributed 31 and
Person_Name
contributed 60, giving a total of 91. Note that, in some cases, the total of the individual field scores may not equal the returned score due to rounding.