Configuring Data Quality with Oracle Enterprise Data Quality

This chapter describes the configuration tasks that you can perform for Oracle Enterprise Data Quality. It covers the following topics:

Overview of Oracle Enterprise Data Quality

The Oracle Enterprise Data Quality suite of products helps organizations by delivering fit-for-purpose data. Oracle Enterprise Data Quality products enable individuals and collaborative teams to identify and resolve problems in underlying data. With Oracle Enterprise Data Quality products, you can identify new opportunities, improve operational efficiency, and more effectively comply with industry or governmental regulation.

The most common data domains in data quality are customer (or more generally, party data which includes suppliers, employees and so on) and product data. Oracle Enterprise Data Quality products address these differences by providing you with the capability to address each.

Installing Oracle Enterprise Data Quality (EDQ)

To install Oracle Enterprise Data Quality, see http://www.oracle.com/webfolder/technetwork/data-quality/edqdocs/oedq_install.pdf.

Note: To support the Oracle Enterprise Data Quality Address Verification with Siebel 8.1.1.11 FP, you must install Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS) version 9.0.4 or later.

Requirements for Installing Oracle Enterprise Data Quality

The following requirements pertain to installation Oracle Enterprise Data Quality (EDQ). For installation instructions, see http://www.oracle.com/technetwork/middleware/oedq/documentation/index.html.

Oracle Enterprise Data Quality (EDQ) Server following the Oracle Enterprise Data Quality (EDQ) Installation Notes from Datanomic.
Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS) must be installed following the instruction from Oracle Enterprise Data Quality Installation Guide.
You must initialize reference data after installing Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS).
You must install Oracle Enterprise Data Quality Address Verification (EDQ-AV) in accordance with instructions with its installation guide. This is required if you want to run the Address Clean process with Oracle Enterprise Data Quality Address Verification (EDQ-AV).

Note: Oracle Enterprise Data Quality Address Verification (EDQ-AV) requires the Global Knowledge Repository, which comprises the data used by Address Verification. This component must be purchased directly from Oracle's partner Loqate ( http://www.loqate.com/). The repository includes its own installer for Windows and other platforms.
Oracle Enterprise Data Quality-Siebel Connector must be installed.
The Java Runtime Environment (JRE) version 6 or later must be installed on the Siebel server.
The version of JRE deployed must match the architecture of the server supported by Siebel CRM. For example, as Siebel supports 32-bit versions of Windows, the 32-bit version of JRE must be installed.

Compatibility of Released Versions of Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS)

The following table lists version compatibility information for Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS).

Table Compatibility for Oracle Enterprise Data Quality Customer Data Services Pack (EDQ-CDS)

Oracle Enterprise Data Quality Customer Data Services Pack	Oracle Enterprise Data Quality	Siebel Connector for Oracle Enterprise Data Quality	EDQ Address Verfication (EDQ-AV)
9.0.1	12.2.1.1.0	9.0.3 to 9.0.5	Any
9.0.2	12.2.1.1.0	9.0.4 to 9.0.5	Any
9.0.3	12.2.1.1.0	9.0.4 to 9.0.5	Any
9.0.4	12.2.1.1.0	9.0.6	12.4.0.0.0 or later

Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification

Siebel Connector calls Oracle Enterprise Data Quality jobs and Web services for real-time data cleansing and matching. A shared staging database is used to pass data between the Siebel CRM application, and Oracle Enterprise Data Quality when batch jobs are run.

This process includes the following tasks:

Configuring the Siebel Server Under Windows

This task is a step in Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification.

To configure Siebel Server under Windows

Copy the dnd.dll file to the Siebel_Server_root\bin directory.
This file is located in the Native sub-folder where you installed the EDQ-Siebel Connector.
Create a file in the Siebel_Server_root\SDQConnector\ directory with the name dnd.parms. This file is used to point to the installation directory on the same server where the Siebel connector runs. The file must contain the following lines:
```
javalib: [location of the JRE]/bin/client/jvm.dll

directory: [the installation directory for the connector]
```
For example:
```
javalib: C:/Program Files/Java/jre1.6/bin/client/jvm.dll

directory: C:/SiebelConnector
```
Note: The Siebel connector is not language specific, so is not installed in a specific language directory.
If detailed trace information on connector requests is required, add the following lines to the dnd.parms file to enable logging:
```
logfile: sdq%05d.log

jlogfile: jsdq%05d.log
```
Extract the siebelconnector.zip file into the installation directory specified in the dnd.parms file (the default location is: Siebel_Server_root\dnd\install.) This installs the following files:
- Jar files for the connector.
- Library jar files needed by the connector.
- Properties files used to store configuration information for the Siebel Connector.
Extract and copy the dnd.properties file from the \OEDQ Customer Data Services Pack 9.0.3\edq-cds-9_0_3_(330)\properties\dnd properties file into the EDQ-Siebel Connector root directory specified in the dnd.params file directory parameter.

Configuring the Siebel Server Under UNIX

This task is a step in Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification.

To configure AIX, Linux and Solaris Siebel Server

Copy the libdnd.so driver file to the Siebel_Server_root/lib directory.
The driver files for each operating system are held in the native subfolder of the siebelconnector.zip file in the Oracle Enterprise Data Quality installation:
- AIX: /native/aix/ppc/ or /native/ppc64, depending on whether a 32-bit or 64-bit version of AIX is in use.
- Linux: /native/linux
- Solaris: /native/sparcv9
Create a file in the Siebel_Server_root/SDQConnector directory with the name dnd.parms. This file will be used to point to the installation directory on the same server where the Siebel connector will run. The file must contain the following lines:
```
javalib: [location of the JRE]/lib/i386/client/libjvm.so

directory: [the installation directory for the connector]
```
For example:
```
javalib: /usr/java/jre1.6/lib/i386/client/libjvm.so

directory: /opt/siebel/dnd/install
```
If detailed trace information on connector requests is required, add the following lines to the dnd.parms file to enable logging:
```
logfile: sdq%05d.log

jlogfile: jsdq%05d.log
```
Extract the siebelconnector.zip file into the installation directory specified in the dnd.parms file (the default being /opt/siebel/dnd/install). This installs the following files:
- Jar files for the connector
- Library jar files needed by the connector
- Properties files used to store configuration information for the Siebel Connector
Extract and copy the dnd.properties file in the EDQ-CDS.zip file over the identically named file in the installation directory.

Configuring the Siebel Server to Use Oracle Enterprise Data Quality

This task is a step in Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification.

To configure the Siebel server to use Oracle Enterprise Data Quality

Log into the Siebel web client as a system administrator.
Navigate to Administration - Server Configuration.
Click the Enterprises link and select the Component Groups tab.
Locate the Data Quality component group and make sure it is enabled.
In the Component Group Assignments view, make sure the component group has been assigned and enabled on the Siebel server.
Click the Parameters tab, and in the Enterprise Parameters list, locate Data Cleaning Type, and set the value to Data Cleansing Type.
Locate the DeDuplication Data Type parameter, and set the value to EDQ.
Click the Servers link, and locate the Data Quality Manager component, then click the Parameters tab.
Make sure the following parameter values are set to EDQ:
- Data Cleansing Type
- DeDuplication Data Type
In the Component Parameters view, make sure the following parameters are set to True:
- Data Cleansing Enable Flag
- DeDuplication Enable Flag
Find the component for the Siebel application which is to use EDQ-CDS, such as the UCM Object Manager (ENU).
Repeat Step 5 through Step 11 for any other components that you want to enable for EDQ-CDS.

Configuring a New Data Quality Vendor

This task is a step in Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification.

To configure a new data quality vendor

Log into the Siebel Web client as a system administrator.
Navigate to Administration - Data Quality.
Click Data Quality Settings and enable the following options:
- Enable DataCleansing = Yes
- Enable DeDuplication = Yes
Add the following options to enable a pop-up dialog box to display possible duplicate records for each new record of a given type when the real-time duplicate prevention service is running.
If these options are not specified, and the real-time duplicate prevention service is running, the check will still be performed, but you must navigate to the Duplicate Accounts, Duplicate Contacts or Duplicate Prospects tab when adding a new record:
- Force User DeDupe - Account = Yes
- Force User DeDupe - Contact = Yes
- Force User DeDupe - List Mgmt = Yes
Click the Third Party Administration link, and in the Vendor list add a new entry with EDQ as the Name value, and dnd at the DLL Name value.

With the newly created vendor record selected, click the BC Vendor Field Mapping tab the enter the following information:

Business Component Name	Operation
Account	Data Cleansing
Account	DeDuplication
Contact	Data Cleansing
Contact	DeDuplication
CUT Address	Data Cleansing
List Mgmt Prospective Contact	Data Cleansing
List Mgmt Prospective Contact	DeDuplication
Personal Address	Data Cleansing

If multiple child entities are enabled, add the following entries:

Business Component Name	Operation
CUT Address	DeDuplication
Personal Address	DeDuplication

In the Field mappings list, set the field mappings for each business component you entered in the BC Operation list. For more information, see Mapping of Vendor Fields to Business Component Fields.

Note: Field mappings can be added or changed as required, provided that they correspond to the configured Web Services in Oracle Enterprise Data Quality (for real-time operations), and the names of the columns in the tables of the staging database (for batch operations).
In the Vendor screen, make sure the EDQ vendor is still selected, then click the Vendor Parameter tab.
Add the parameters listed in Configuring Vendor Parameters.

Disabling Informatica Identity Resolution Synchronization

This task is a step in Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification.

Note: This task is required only for users of Oracle Data Quality Matching Server, which uses Informatica Identity Resolution software. This task allows you to disable synchronization used by Informatica Identity Resolution.

To disable Informatica Identity Resolution synchronization

Navigate to Administration - Runtime Events, then Action Sets.
In the Action Sets view, query for DQ Sync in the Action Sets field.
Deselect the Active flag to inactivate all related action sets.
Navigate to Administration - Data Quality, then Data Quality Settings.
In the Data Quality Settings view, set the Enable DQ Sync parameter value to No.

Process of Configuring the Staging Database for Oracle Enterprise Data Quality

The staging database is used by the Siebel Connector as a staging area for batch matching jobs on the master data.

Note: In Oracle Customer Hub (UCM), batch matching is normally performed using the Oracle Enterprise Data Quality real-time matching services, so that all connected source systems are considered during matching. However, for both Siebel CRM and Oracle Customer Hub (UCM) the Siebel Data Quality Manager service can run batch jobs on the master data only.

Candidates for matching are written to a table to be read by an Oracle Enterprise Data Quality job. Oracle Enterprise Data Quality then exports duplicates from the matching process to a table to be read back into the Siebel database.

This process includes the following tasks:

Creating Tables

This task is a step in Process of Configuring the Staging Database for Oracle Enterprise Data Quality.

The sql folder in the root of the EDQ-CDS distribution contains six scripts that can be run to create tables as follows:

siebel_account_candidates.sql creates the SIEBEL_ACCOUNT_CANDIDATES table.
siebel_account_matches.sql creates the SIEBEL_ACCOUNT_MATCHES table.
siebel_contact_candidates.sql creates the SIEBEL_CONTACT_CANDIDATES table.
siebel_contact_matches.sql creates the SIEBEL_CONTACT_MATCHES table.
siebel_prospect_candidates.sql creates the SIEBEL_PROSPECT_CANDIDATES table.
siebel_prospect_matches.sql creates the SIEBEL_PROSPECT_MATCHES table.

Configuring Connections

This task is a step in Process of Configuring the Staging Database for Oracle Enterprise Data Quality.

Both the Siebel Connector and EDQ-CDS itself must connect to the staging database to read and write to the Candidate and Match tables when processing batch jobs. These tables can be created in any database and schema, but by default the connection is configured for an Oracle database.

The connection details are specified in the dnd.properties file, see Process of Configuring Oracle Enterprise Data Quality for Matching and Address Verification. To configure these, open the dnd.properties file and edit the six parameters in the Staging database connection details section. These parameters control the database host, port, credentials and other settings used to connect to the staging database.

Process of Configuring Oracle Enterprise Data Quality to Find Contact and Address Matches

By default, Oracle Enterprise Data Quality does not find matches with the following search constraints: Contact First and Last Name only or Account Name and Site only. A match is made only when addresses are supplied for the contact or account. To find matches with Contact First and Last Name only or Account Name and Site only without addresses, you must activate a rule within Oracle Enterprise Data Quality.

To configure Oracle Enterprise Data Quality, perform the following tasks:

Activating Oracle Enterprise Data Quality Rules

By default, Oracle Enterprise Data Quality does not find matches with the following search constraints: Contact First and Last Name only or Account Name and Site only. A match is only made when addresses are supplied for the contact or account. To find matches with Contact First and Last Name only or Account Name and Site only without addresses, you must activate a rule within Oracle Enterprise Data Quality.

This task is a step in Process of Configuring Oracle Enterprise Data Quality to Find Contact and Address Matches.

To activate Oracle Enterprise Data Quality rules

In Oracle Enterprise Data Quality, navigate to Director, then open the following project: EDQ-CDS.
Open the following processes:
- Individual Match - Real-Time
- Individual Match - Batch
For each process, open the Match Individuals step, then click the Match icon to open the match rules.
Select the checkbox to activate the following rules:
- [I010U] Script full name exact; no address. Set the value to 65.
- [I020U] Name exact; no address. Set the value to 65.
Repeat each Step 2 through Step 4 for the following processes:
- Account Entity Match - Real-Time
- Entity Match - Batch
The rule returns a score of only 60 there is a match on Account Name and Site fields.
Save the project.
Restart Oracle Enterprise Data Quality services and the Real-time START ALL jobs.

Configuring Siebel Match Threshold Settings

Because the Oracle Enterprise Data Quality rule now has matching scores of 60 and 65, you must change the Siebel application Match Threshold setting so that your Siebel application can identify the new matches.

This task is a step in Process of Configuring Oracle Enterprise Data Quality to Find Contact and Address Matches.

To change the Siebel match threshold setting

Navigate to Administration - Data Quality, then Data Quality Settings.
Change the Match Threshold setting from the default value of 70 to 50.
Restart Siebel Server services.

Vendor Parameters for Oracle Enterprise Data Quality

The following table lists parameters which allow you to enable and control third party data quality vendors.

Table Parameters for enabling third-party vendors in Oracle Enterprise Data Quality

Name	Value
Key Generation Process by Third Party	The default value is No. This parameter should only be set to Yes if external key generation is supported by the installed version of Siebel. For more information, see Oracle Customer Hub (UCM) Master Data Management Reference. If this parameter is set to No, the Query Expression parameters and the Token Expression parameters must also be set. For more information, see Match Key Generation Using the Universal Connector with Third-Party Data Quality Vendors.
Support Multiple Child Entities Deduplication	The default value is Yes. Note: If this parameter is set to Yes, the CUT and Personal Address DeDup Record Type parameters, which are displayed in this table, are enabled. Also note that this parameter should only be set to Yes if Multiple Child Entities are supported by the installed version of Siebel CRM.
Account DataCleanse Record Type	accountclean.
Account DeDup Record Type	accountmatch.
CUT Address DataCleanse Record Type	addressclean.
CUT Address DeDup Record Type	addressmatch. Note: Only used if multiple child entities are enabled.
Personal Address DataCleanse Record Type	addressclean.
Personal Address DeDup Record Type	addressmatch. Note: Only used if multiple child entities are enabled.
Contact DataCleanse Record Type	contactclean.
Contact DeDup Record Type	contactmatch.
List Mgmt Prospective Contact DataCleanse Record Type	prospectclean.
List Mgmt Prospective Contact DeDup Record Type	prospectmatch.
Batch Max Num of Records	200
Parameter 1	"session", "clusterlevel", "3". Note: There must be a space after each comma.
Parameter 2	"session", "matchthreshold", "50". Note: There must be a space after each comma.

Oracle Enterprise Data Quality Field Mappings for Siebel Business Components

The following field mappings work with the provided templates for data quality services, and the provided example SQL for creating the staging database.

Note the following:

All fields in these business components are drop-down fields, with the exception of Id. This field must be completed manually.
If Oracle Enterprise Data Quality Address Verification is used for address cleaning:
- The State field mapping only functions correctly if all addresses in the Siebel database are US addresses. If the addresses are not all US addresses, do not map this field.
- The Country field mapping may not function correctly if the Siebel CRM field is restricted to fixed list of values that do not exactly match those returned by the cleaning process.

This topic contains information on the following field mappings:

Account - Data Cleansing

The following table lists the mapped fields for data cleaning of the Account business component.

Table Mapped fields for Account data cleansing

Business Component Field	Mapped Field
Id	entityid
Language Code	languages
Name	name
Location	subname
Main Phone Number	phone
Tax ID Number	taxnumber
Home Page	website
VAT registration number	vatnumber

Account - DeDuplication

The following table lists the mapped fields for deduplication of the Account business component.

Table Mapped fields for Account deduplication

Business Component Field	Mapped Field
Id	entityid
Language Code	languages
Name	name
Location	subname
Main Phone Number	phone
Tax ID Number	taxnumber
VAT registration number	vatnumber
Home Page	website

The following table lists the mapped fields for deduplication of the Account business component when the Multiple Child Entities parameter is disabled.

Table Mapped fields for Account deduplication when Multiple Child Entities is disabled

Business Component Field	Mapped Field
Primary Account Street Address	address1
Primary Account Address Street Address2	address2
Primary Account Address Street Address3	address3
Primary Account City	city
Primary Account State	adminarea
Primary Account Postal Code	postalcode
Primary Account Country	country

CUT Address - Data Cleansing

The following table lists the mapped fields for data cleansing of the CUT Address business component.

Table Mapped fields for CUT Address data cleansing

Business Component Field	Mapped Field
Id	addressid
Street Address	address1
Street Address 2	address2
Street Address 3	address3
Street Address 4	address4
City	city
Neighborhood	doubledependentlocality
Metro Area	dependentlocality
County	subadminarea
State	adminarea
Postal Code	postalcode
Country	country

CUT Address - DeDuplication

The following table lists the mapped fields for deduplication of the CUT Address business component.

Note: This business component is only used if the Multiple Child Entities parameter is enabled.

Table Mapped fields for CUT Address deduplication

Business Component Field	Mapped Field
Street Address	address1
Street Address 2	address2
City	city
State	adminarea
Postal Code	postalcode
Country	country

Personal Address - Data Cleansing

The following table lists the mapped fields for data cleansing of the Personal Address business component.

Table Mapped fields for Personal Address data cleansing

Business Component Field	Mapped Field
Id	addressid
Street Address	address1
Street Address 2	address2
Street Address 3	address3
Street Address 4	address4
City	city
Neighborhood	doubledependentlocality
Metro Area	dependentlocality
County	subadminarea
State	adminarea
Postal Code	postalcode
Country	country

Personal Address - DeDuplication

The following table lists the mapped fields for deduplication of the Personal Address business component.

Note: This business component is only used if the Multiple Child Entities parameter is enabled.

Table Mapped fields for Personal Address deduplication

Business Component Field	Mapped Field
Street Address	address1
Street Address 2	address2
City	city
State	adminarea
Postal Code	postalcode
Country	country

Contact - Data Cleansing

The following table lists the mapped fields for data cleansing the Contact business component.

Table Mapped fields for Contact data cleansing

Business Component Field	Mapped Field
Id	individualid
Language Code	languages
M/M	title
M/F	gender
First Name	firstname
Middle Name	middlename
Last Name	lastname
Home Phone #	homephone
Work Phone #	workphone
Fax Phone #	faxphone
Cellular Phone #	mobilephone
Email Address	email
Job Title	jobtitle
Social Security Number	taxnumber
Birth Date	dob

Contact - DeDuplication

The following table lists the mapped fields for deduplication of the Contact business component.

Table Mapped fields for Contact deduplication

Business Component Field	Mapped Field
Id	individualid
Language Code	languages
M/M	title
M/F	gender
First Name	firstname
Middle Name	middlename
Last Name	lastname
Home Phone #	homephone
Work Phone #	workphone
Fax Phone #	faxphone
Cellular Phone #	mobilephone
Email Address	email
Job Title	jobtitle
Social Security Number	taxnumber
Birth Date	dob
Primary Account Name	accountname

The following table lists the mapped fields for deduplication of the Contact business component when the Multiple Child Entities parameter is disabled.

Table Mapped fields for Contact deduplication when Multiple Child Entities is disabled

Business Component Field	Mapped Field
Primary Street Address	address1
Primary Address Street Address2	address2
Primary Address Street Address3	address3
Primary City	city
Primary State	adminarea
Primary Postal Code	postalcode
Primary Country	country

List Mgmt Prospective Contact - Data Cleansing

The following table lists the mapped fields for data cleansing the List Mgmt Prospective Contact business component.

Table Mapped fields for List Mgmt Prospective Contact data cleansing

Business Component Field	Mapped Field
Id	individualid
M/M	title
M/F	gender
First Name	firstname
Middle Name	middlename
Last Name	lastname
Home Phone #	homephone
Work Phone #	workphone
Fax Phone #	faxphone
Cellular Phone #	mobilephone
Email Address	email
Job Title	jobtitle
Social Security Number	taxnumber
Birth Date	dob

List Mgmt Prospective Contact - DeDuplication

The following table lists the mapped address fields for deduplication of the List Mgmt Prospective Contact business component.

Note: List Mgmt Prospective Contact has just one address. You can, however, include the address fields in the mapped fields for deduplication of List Mgmt Prospective Contact, regardless of whether the Multiple Child Entities parameter are enabled or not.

Table Mapped fields for List Mgmt Prospective Contact deduplication

Business Component Field	Mapped Field
Id	individualid
M/M	title
M/F	gender
First Name	firstname
Middle Name	middlename
Last Name	lastname
City	city
Country	country
Postal Code	postalcode
State	adminarea
Street Address	address1
Street Address 2	address2
Home Phone #	homephone
Work Phone #	workphone
Fax Phone #	faxphone
Cellular Phone #	mobilephone
Email Address	email
Job Title	jobtitle
Account	accountname
Social Security Number	taxnumber
Birth Date	dob

Configuring Third-Party Key Generation for Oracle Enterprise Data Quality

With the Siebel Data Quality (SDQ) Universal Connector framework the dedup keys used in the candidate selection process are generated by Siebel CRM using a simple token expression. To enable more functionality in key generation you can enable external key generation through a third-party data quality vendor.

To enable third-party key generation for Oracle Enterprise Data Quality, set the Key Generation Process by Third Party parameter to Yes as shown in Vendor Parameters for Oracle Enterprise Data Quality.

For implementation, the connector API is sdq_get_dedup_key.

Enabling Support of Multiple Child Entity Deduplication for Oracle Enterprise Data Quality

When using third-party data quality vendors, you can enable Oracle Enterprise Data Quality to perform deduplication on multiple child entities of parent objects, such as Personal Address for the Contact object.

To enable multiple child entity deduplication for Oracle Enterprise Data Quality, set the Support Multiple Child Entities Deduplication parameter to Yes as shown in the following image. For an example of the data format, see Data Format for Deduplication of Multiple Child Entities

Data Format for Deduplication of Multiple Child Entities

To enhance the functionality of Siebel Universal Connector and a third party external key generating vendor, the Siebel Universal Connector API supports hierarchical data format. For example, instead of making multiple calls to the third party vendor for multiple address deduplication, the hierarchical data format sends all child addresses in one payload, thus enabling the connector to make just one call.

The following example shows the data format with the ListOf container used for each child entity.

Note the following:

The Name field is derived from the parent business component.
The ListOf container is used for the Address child object.
The Street field is a join field derived from the child business component.

<Data>
   <DriverRecord>
         <Name>…</Name>
         ...
         <ListOfAddress>
                <Address>
                      <Street>…</Street>
                      ...
                </Address1>
                <Address2>
                      <Street>…</Street>
                      ...
                </Address2>
                ...
                <AddressN>
                      <Street>…</Street>
                      ...
                </AddressN>
         </ListOfAddress>
   </DriverRecord>
   <CandidateRecord>
   </CandidateRecord>
</Data>

The following example shows the data format for the Contact object.

<Data>
   <DriverRecord>
         <Contact.Id>12-12ABC1</Contact.Id>
         <FirstName>abc</FirstName>
         <LastName>abc</LastName>
         <ListOfAccount>
                   <Account>
                            <Name>A. K. Parker Inc</Name>
                            <Location>San Mateo</Location>
                   </Account>
                   <Account>
                            <Name>Hibbing Manufacturing</Name>
                            <Location>San Jose</Location>
                   </Account>
         </ListOfAccount>
         <ListOfPersonalAddress>
                   <PersonalAddress>
                            <City>York1</City>
                            <Country>USA</Country>
                            <Zip>78201</Zip>
                            <State>AL</State>
                            <Street> 1200 Davisville Road</Street>
                   </PersonalAddress>
         </ListOfPersonalAddress>
   </DriverRecord>
   <CandidateRecord>
         <Contact.Id>12-BCDD</Contact.Id>
         <FirstName>abc1</FirstName>
         <LastName>abc1</LastName>
         <ListOfAccount>
                   <Account>
                            <Name>A. K. Parker Inc</Name>
                            <Location>San Mateo</Location>
                   </Account>
         </ListOfAccount>
         <ListOfPersonalAddress>
                   <PersonalAddress>
                            <City>York1</City>
                            <Country>USA</Country>
                            <Zip>78201</Zip>
                            <State>AL</State>
                            <Street> 1200 Davisville Road</Street>
                   </PersonalAddress>
         </ListOfPersonalAddress>
   </CandidateRecord>
   <CandidateRecord>
         <Contact.Id>12-BCEE</Contact.Id>
         <FirstName>abc2</FirstName>
         <LastName>abc2</LastName>
         <ListOfAccount> 
         </ListOfAccount>  
                <ListOfPersonalAddress>
                   <PersonalAddress>
                            <City>York3</City>
                            <Country>USA</Country>
                            <Zip>78203</Zip>
                            <State>AL</State>
                            <Street> 1203 Davisville Road</Street>
                   </PersonalAddress>
                </ListOfPersonalAddress>
   </CandidateRecord>
</Data>

The following example shows the data format for the Account object.

<Data>
   <DriverRecord>
         <Account.Id>12-ABCC</Account.Id>
         <Name>Hibbing Manufacturing</Name>
         <Location>San Jose</Location>
         <ListOfCUTAddress>
                   <CUTAddress>
                            <City>York1</City>
                            <Country>USA</Country>
                            <Zip>78201</Zip>
                            <State>AL</State>
                            <Street> 1200 Davisville Road</Street>
                   </CUTAddress>
         </ListOfCUTAddress>
   </DriverRecord>
   <CandidateRecord>
         <Account.Id>12-AAAA</Account.Id>
         <Name>Hibbing Manufacturing1</Name>
         <Location>San Jose</Location>
         <ListOfCUTAddress>
                   <CUTAddress>
                            <City>York1</City> 
                            <Country>USA</Country>
                            <Zip>78201</Zip>
                            <State>AL</State>
                            <Street> 1200 Davisville Road</Street>
                   </CUTAddress>
                   <CUTAddress>
                            <City>York2</City>
                            <Country>USA</Country>
                            <Zip>78202</Zip>
                            <State>AL</State>
                            <Street> 1201 Davisville Road</Street>
                   </CUTAddress>
         </ListOfCUTAddress>
   </CandidateRecord>
   <CandidateRecord> 
         <Account.Id>12-BBBB</Account.Id>
         <Name>Hibbing Manufacturing2</Name>
         <Location>San Francisco</Location>
   </CandidateRecord>
   <CandidateRecord>
         <Account.Id>12-CCCC</Account.Id>
         <Name>Hibbing Manufacturing3</Name>
         <Location>San Mateo</Location>
         <ListOfCUTAddress>
                   <CUTAddress>
                            <City>York1</City>
                            <Country>USA</Country>
                            <Zip>78201</Zip>
                            <State>AL</State>
                            <Street> 1200 Davisville Road</Street>
                   </CUTAddress>
         </ListOfCUTAddress>
   </CandidateRecord>
</Data>

Overview of Configuring Oracle Enterprise Data Quality for Deduplication of Child Entities

Oracle Enterprise Data Quality (EDQ) can be configured to support the deduplication of a parent object’s child entities. The steps to do this depend on the complexity of the child entity you want to use. When multiple child entity matching is enabled hierarchical data from Siebel CRM is transformed by the Siebel Connector into a flat record structure so that EDQ can match parent records (such as Contacts, Prospects, or Accounts) using the details of the child records (such as Names, Addresses, Email Addresses, Alternative Phone Numbers, and so on). Siebel Connector can be configured to handle child entities either as scalar (single field) or non-scalar (multi-field) objects.

Scalar groups are used for Siebel business components comprising one field, or for which only one field is mapped in the Data Quality field mappings in the Siebel database. The connector prepares the values for such groups into a simple delimited list of values in a single input attribute for EDQ.

Non-scalar groups are used for more complex child entities comprising multiple fields, such as names and addresses. In this case, the connector prepares multiple records for the same parent to submit to EDQ for matching purposes, each with different child data.

The following is an overview of the steps required to configure Enterprise Data Quality to process additional child entity types. For more information about extending the matching configuration, see Oracle® Enterprise Data Quality Customer Data Services Pack Customization Guide 11g Release 1 (11.1.1.7) on Oracle Technology Network (OTN).

To extend the matching configuration in Oracle Data Quality (EDQ)

In Oracle Data Quality (EDQ), extend the EDQ-CDS project to provide configuration logic to implement deduplication for the child entity type as follows:
1. Edit the Web service definition (corresponding to the appropriate parent object type) to include the new attributes for the child entity type.
2. Edit the candidates snapshot to include the new attributes.
3. Edit the candidates data interface to include the new attributes.
4. Edit the data interface mapping for the snapshot to include mappings for the new attributes.
5. Edit the data interface mapping for the Web service to include mappings for the new attributes.
6. Edit the cluster process to implement any new cluster key generation logic required for the child entity type.
7. Edit the match process to implement the new matching logic for the child entity type.
Modify EDQ Siebel Connector dnd.properties file as follows:
1. Edit the Multiple child entity groupings section to define the child entity as either a scalar or non-scalar group as appropriate.
2. Restart the Siebel Server to apply the changes.

Process of Configuring Additional Child Entities for Deduplication

You can enable or disable support for multiple child entity deduplication with Oracle Enterprise Data Quality (EDQ). By default this feature supports the following child entities: CUT Address, and Personal Address. To enable additional child entities for deduplication, you must first modify the EDQ Siebel Connector and EDQ configuration (including processes, Web services, and matching rules), then perform the steps in this process. To modify the EDQ Connector, see Overview of Configuring Oracle Enterprise Data Quality for Deduplication of Child Entities, and Oracle® Enterprise Data Quality Installation and Upgrade Guide Release 11g R1 (11.1.1.7) on Oracle Technology Network (OTN), and Oracle® Enterprise Data Quality Customer Data Services Pack Customization Guide 11g Release 1 (11.1.1.7) on Oracle Technology Network (OTN).

Note: As an example, this process uses Communication Address (Alternate Email Address) as the child entity for deduplication of the Contact object.

To configure additional child entities for deduplication, perform the following tasks:

Modifying the Data Quality Integration Object for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to modify the Data Quality integration object for additional child entity deduplication.

To modify an integration object for additional child entity deduplication

In the Web Tools Object Explorer, select Integration Object.
Query for the integration object you want to modify. In this example, query for Contact Dedup IO.

Note: The Contact Dedup integration object is used by Oracle Enterprise Data Quality to query for contacts and all child components when performing deduplication. Modify Account Dedup IO if you want to include an additional child entity deduplication for the Account object.

In the Integration Component view, add the necessary integration component as shown in the following table.

Note: For this example, you must add the Communication Address integration component.

Name	External Name	External Name Context	Parent Integration Object Component	XML Container Element	XML Tag
Contact	Contact	Contact	Not applicable	Not applicable	Contact
Personal Address	Personal Address	Contact_INS Personal Address	Contact	ListOfPersonalAddress	PersonalAddress
Communication Address	Communication Address	Contact_Communication Address	Contact	ListOfCommunicationAddress	CommunicationAddress

Select the Communication Address integration component, then in the Integration Component Fields view, add the necessary integration component fields as shown in the following table:

Name	Data Type	Length	External Name	XML Style	XML Tag
Id	DTYPE_ID	30	Communication AddressId	Element	Contact
Address	DTYPE_TEXT	100	Alternate Email Address	Element	AlternateEmailAddress
Medium Type	DTYPE_TEXT	30	Communication Address Medium Type	Element	CommunicationAddressMediumType
Name	DTYPE_TEXT	50	Communication Address Name	Element	CommunicationAddressName
User Type	DTYPE_TEXT	30	Communication Address Use Type	Element	CommunicationAddressUse Type

Modifying the Data Quality Business Service for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to modify the Data Quality business service for additional child entity deduplication.

To modify the Data Quality business service for additional child entity deduplication

In the Web Tools Object Explorer, select Business Service
Query for the Data Quality business service: DeDuplication.
In the Business Service User Props view, select the necessary user property.
For example, select Contact Child Dedup Object Name and add Communication Address to the list as follows:
- Name: Contact Child Dedup Object Name
- Value: Personal Address;Communication Address.
Note: List all child Dedup objects for Contact and separate each object entry with a semicolon (;).

Deploying the Integration Object for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to deploy the modified repository objects for additional child entity deduplication.

To deploy the integration object for additional child entity deduplication

Stop the Siebel Server and deliver the modified integration object and business service to your Siebel runtime repository, then deploy the integration object by doing the following:
1. Select the modified integration object in Web Tools, right-click and choose Undeploy, then step off the record to retain the changes.
2. Right-click again on the integration object and choose Deploy to Runtime Database, then step off the record.
Start the Siebel Server.

Mapping the Third Party Vendor Field for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to map the third party vendor field for additional child entity deduplication.

To map the third party vendor field for additional child entity deduplication

In Oracle Customer Hub (UCM), navigate to Administration - Data Quality, then Third Party Administration.
From the Third Party Administration screen, select EDQ in the Vendor view.
Click the BC Vendor Field Mapping tab, then in the BC Vendor Field Mapping view, add a new entry for the Business Component, and Operation fields. For this example enter the following:
- Business Component: Communication Address
- Operation: Deduplication
In the Field Mappings view, enter the correct business component field. For this example, enter the following:
- Business Component Field: Address
- Mapped Field: email
For a list of all vendor field mappings, see Oracle Enterprise Data Quality Field Mappings for Siebel Business Components.

Setting the Third Party Vendor Parameter for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to set the third party vendor parameter for additional child entity deduplication.

To set the third party vendor parameter for additional child entity deduplication

In Oracle Customer Hub (UCM), navigate to Administration - Data Quality, then Third Party Administration.
From the Third Party Administration screen, select EDQ in the Vendor view.
Click on the Vendor Parameter tab, and add a new record to specify the DeDup Record Type for the new child component. For this example, enter the following:
- Vendor Parameter Name: Communication Address DeDup Record Type
- Value: emailmatch
For a list of all vendor parameters, see Vendor Parameters for Oracle Enterprise Data Quality.

Configuring Data Maps for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication.

To use a new child entity for matching in Oracle Customer Hub (UCM) processes, you must add the new integration component and integration component fields to Oracle Customer Hub (UCM) integration objects (such as CIFContact, SwiPersonIO, or UCMContactSourceDataAndHistory). Additionally, to allow Oracle Customer Hub (UCM) to send the new fields to Oracle Enterprise Data Quality for matching, you must map the new fields from the Oracle Customer Hub (UCM) integration object to the Oracle Enterprise Data Quality integration object.

To perform the following steps you must have performed all preceding steps in this process to modify all required objects to include the new child integration component and integration component fields, and also the modified integration objects must have been redeployed to the Siebel run-time database.

Use this task to configure data maps for additional child entity deduplication.

To configure data maps for additional child entity deduplication

In Oracle Customer Hub (UCM), and navigate to Administration - Integration, then Data Map Editor.
In the Integration Object Map view, query for the required integration object map.
For this example, use the following query: UCM Contact*DQ IO.
Modify the required maps.
For this example, modify the following maps for Contact Deduplication:
- UCM Contact CIF IO to DQ IO (used by the UCM Real-time Process).
- UCM Contact SDH IO to DQ IO (used by the UCM Batch Process).
- UCM Contact SOAP IO to DQ IO (used by the UCM Soap Process).
- UCM Contact Match IO to DQ IO (used by the Person Match Web Service).

For each data map in this example, you must use the Integration Component Map view to add the Contact_Communication Address integration component. See the following table:

Name	Source Component Name	Target Component Name	Source Search Specficiation
Contact	Contact	Contact	Not applicable
Contact_INS Personal Address	Contact INS Personal Address	Personal Address	Not applicable
Contact Account	Contact Account	Contact	IsPrimaryMVG = ‘Y’
Contact_Communication Address	Contact Communication Address	Communication Address	Not applicable

Select the newly created integration component map, Contact_Communication Address, then in the Integration Field Map view add the mapping for an alternate email address using the following example:
- Source Expression: [Alternate Email Address]
- Target Field Name: Address
Repeat Step 4 and Step 5 for each additional data map.

Configuring Oracle Enterprise Data Quality-Siebel Connector for Additional Child Entity Deduplication

This task is a step in Process of Configuring Additional Child Entities for Deduplication. Use this task to configure Oracle Enterprise Data Quality-Siebel Connector for additional child entity deduplication.

For more information on the Scalar groups area of the dnd.properties file, see Overview of Configuring Oracle Enterprise Data Quality for Deduplication of Child Entities.

To configure the Oracle Enterprise Data Quality-Siebel Connector for additional child entity deduplication

Locate the dnd.properties file for Oracle Enterprise Data Quality-Siebel Connector.
In the properties file, locate the Multiple child entity groupings section, and under the Scalar groups to be stored as delimited lists in a single field sub-section, add the following scalar group:
- group.email. Add a property value of CommunicationAddress.
- group.email.concat. Add a property value of email.
- group.email.delimiter. Add a property value of |.
With these settings, Oracle Enterprise Data Quality-Siebel Connector appends all alternate email addresses in the CommunicationAddress child entity and sends them as a single field (in an email attribute) to the Oracle Enterprise Data Quality (EDQ) server for matching.

Note: Support for matching against multiple child entities was added in EDQ-CDS version 9.0.2 but it is recommended that the latest compatible version of the software is used. See the Enterprise Data Quality Certification Matrix for details about compatible software versions: http://docs.oracle.com/cd/E48549_01/doc.11117/e40736/cdsis.htm#sthref4.

Sample Message from Siebel CRM to EDQ Siebel Connector for Additional Child Entity Deduplication

The following is a sample message with multiple email addresses sent from Siebel CRM. Note that there are is email address at the parent contact level (DriverRecord). The list of alternate email addresses are sent in the child entity (Communication Address) which maps to the Email attribute in EDQ.

<Data>
   <DriverRecord>
         <firstname>Mike</firstname>
         <lastname>Jones</lasttname> 
         <workphone>1234567890</workphone>
         <ListofPersonalAddress>
            <PersonalAddress>
                <address1>100 First Street</address1>
                <city>San Jose</city>
                <state>CA</state>
                <country>USA</country>
            </PersonalAddress>
         <ListofCommunicationAddress>
            <CommunicationAddress>
                <email>mike@yahoo.com</email>
                <usage>Personal</usage>
            </CommunicationAddress>
            <CommunicationAddress>
                <email>mike@stanford.edu</email>
                <usage>Campus</usage>
            </CommunicationAddress>
            <CommunicationAddress>
                <email>mike@oracle.com</email>
                <usage>Business</usage>
            </CommunicationAddress>
         </ListofCommunicationAddress>
   </DriverRecord>
</Data>

Overview of Configuring Social Media Child Objects for Matching

With Siebel Innovation Pack 2014 you can enable the Contact object child entity "SM Contact Social Profile" for matching in Oracle Customer Hub (UCM). Configuration tasks include:

Prerequisite

To perform this process, you must first have completed the following task:

Configure the EDQ Connector, as described in: Overview of Configuring Oracle Enterprise Data Quality for Deduplication of Child Entities

Then perform the tasks listed in the process.

To configure additional child entities for deduplication, perform the following tasks:

Sample Message from Siebel CRM to EDQ Siebel Connector for Social Media Child Object Matching

The following is a sample message with multiple Contact Social Profiles sent from Siebel CRM to EDQ Siebel Connector for Contact matching.

<Data>
   <DriverRecord>
         <Contact.Id>12-12ABC1</ContactId>
         <firstname>Mike</firsttname> 
         <lastname>Jones</lasttname> 
         <ListofSMContactSocialProfile>
            <SMContactSocialProfile>
                <Author>Mike Jones</Author>
                <Community>Facebook</Community>
            </SMContactSocialProfile> 
            <SMContactSocialProfile>
                <Author>Mike Jones</Author>
                <Community>Google</Community>
            </SMContactSocialProfile>
            <SMContactSocialProfile>
                <Author>jones</Author>
                <Community>Instagram</Community>
            </SMContactSocialProfile>
            <SMContactSocialProfile>
                <Author>@jones</Author>
                <Community>Twitter</Community>
            </SMContactSocialProfile>
            <SMContactSocialProfile>
                <Author>Mike Jones</Author>
                <Community>YouTube</Community>
            </SMContactSocialProfile>
         </ListofSMContactSocialProfile>
   </DriverRecord>
</Data>