6 Match Types and Field Names

When you create an Oracle Healthcare Master Person Index (OHMPI) application, you can select a Match Type for each field in the OHMPI wizard. Each match type defines a different type of standardization, normalization, phonetic encoding, and matching logic in the mefa.xml file. The following sections describe each match type you can specify in the wizard and how each affects the logic in the mefa.xml file.

This chapter includes the following sections:

Match Types and Standardization Types
Oracle Match Engine Match Types

Match Types and Standardization Types

For each field that will be used for matching in the OHMPI application, you can select a match type in the wizard. When you select a match type for a field, Oracle Healthcare Master Person Index automatically adds that field to the match string in the mefa.xml file and, in many cases, generates additional fields in the object structure that are not visible on the wizard. These fields are used for searching and matching and they should not be modified.

If new fields are generated, they are automatically incorporated into the configuration files and the database script that creates the master person index tables. These fields store standardized, normalized, or phonetic versions of the field, depending on the type of matching you choose. In addition, these fields are assigned a match type in the match string in the mefa.xml file. They might also be defined for standardization in the mefa.xml file, in which case they will also be assigned a standardization type. The types described in the next section pertain to the OHMPI Match Engine only.

Note:

The match types specified in the mefa.xml file for the fields in the match string are not always the same as the match types you specify in the wizard. The match types you can specify from the Configuration Editor more closely match those that are shown in the mefs.xml file. This discussion pertains more to the match types on the wizard. For more information, see Oracle Healthcare Master Person Index Match Engine Reference.

Oracle Match Engine Match Types

The Oracle Healthcare Master Person Index wizard match types for the OHMPI Match Engine fall into four primary categories.

Person Match Types
BusinessName Match Types
Address Match Types
Miscellaneous Match Types

The actual standardization and match types entered into the mefa.xml file vary for each match type you select in the wizard. The match and standardization types for each type of field are listed in the following descriptions. The match types entered into the mefa.xml file correspond to the match types defined in the match configuration file, MatchConfigFile.cfg.

Person Match Types

The Person match types include PersonLastName and PersonFirstName. These match types are used to normalize and phonetically encode name fields for person matching. For each field with one of these match types, the wizard adds two fields to the object structure for phonetic and standardized versions. If you specify a field with a person match type for blocking in the wizard, the phonetic version of the name is automatically added to the blocking query. The following fields are created when you specify one of the Person match types for a field (field_name refers to the name of the field specified for Person matching).

field_name_Std - This field contains the normalized version of the name.
field_name_Phon - This field contains the phonetic version of the name.

The corresponding standardization and match types in the mefa.xml file are listed in Table 6-1.

Table 6-1 Person Name Standardization and Match Types

MPI Wizard Match Type	Match Field File Standardization Type	mefa.xml Match Type
PersonLastName	PersonName	LastName
PersonFirstName	PersonName	FirstName

BusinessName Match Types

The BusinessName match type is designed to help parse, normalize, and phonetically encode a business name. BusinessName matching adds several fields to the object structure and to the match string. If you specify a business name field for blocking, each parsed business name field is added to the blocking query. The corresponding standardization type in the mefa.xml file for all fields selected for BusinessName matching is also BusinessName. The actual match type assigned to each field varies depending on the type of information in each field.

Table 6-2 lists the fields created when you select the BusinessName match type for a field along with their corresponding match types in the mefa.xml file (field_name refers to the name of the field selected for BusinessName matching).

Note:

Only specify this type of matching for one business name field; otherwise, the wizard will create duplicate entries in the object structure. If more than one field contains the business name, you can add those fields to the standardization structure in the mefa.xml file after the wizard creates the configuration files.

Table 6-2 BusinessName Match Types

Field Name	Description	Added to Match String	mefa.xml Match Type
field_name_Name	The parsed and normalized version of the business name.	Yes	PrimaryName
field_name_NamePhon	The phonetic version of the business name.	No
field_name_OrgType	The parsed organization type of the business name.	Yes	OrgTypeKeyword
field_name_AssocType	The association type for the business.	Yes	AssocTypeKeyword
field_name_Industry	The name of the industry for the business.	Yes	IndustryTypeKeyword
field_name_Sector	The name of the industry sector (industries are a subset of sectors).	Yes	IndustrySectorList
field_name_Alias	An alias for the business name.	No	-
field_name_Url	The web site URL for the business.	Yes	Url

Address Match Types

The Address match type is designed to help parse, normalize, and phonetically encode an address for matching or standardizing address information. Address matching adds several fields to the object structure and to the match string. If you specify an address field for blocking, the parsed fields are added to the blocking query. The corresponding standardization type for fields selected for Address matching is Address. The actual match type assigned to each field varies depending on the type of information in each field.

The fields created when you select the Address match type for a field are listed in Table 6-3, along with their corresponding match types in the mefa.xml file (field_name refers to the name of the field selected for Address matching).

Note:

Only specify this type of matching for one street address field; otherwise, the wizard will create duplicate entries in the object structure. If more than one field contains the street address, you can define the additional fields in the standardization structure in the mefa.xml file after the wizard creates the configuration files.

Table 6-3 Address Match Types

Field Name	Description	Added to Match String?	mefa.xml Match Type
field_name_HouseNo	The parsed street number of the address.	Yes	HouseNumber
field_name_StDir	The parsed and normalized street direction of the address.	Yes	StreetDir
field_name_StName	The parsed and normalized street name of the address.	Yes	StreetName
field_name_StPhon	The phonetic version of the street name.	No	-
field_name_StType	The parsed and normalized street type of the address, such as Boulevard, Street, Drive, and so on.	Yes	StreetType

If you want to search on street addresses but do not want to use these fields for matching, select the Address match type for only one street address field in the wizard. When the wizard is complete, you can remove the address fields from the match string in the mefa.xml file.

Miscellaneous Match Types

Several additional match types are defined in the wizard for the OHMPI Match Engine. These match types are used to indicate matching on string, date, or number fields other than those described above or to indicate matching on a field that contains a single character (such as the gender field, which might accept F for female or M for male). These match types do not define standardization for the specified field and do not add any fields to the object structure. If you specify one of these match types for a field in the wizard, the field is added to the match string with a match type of String, Date, Number, or Char.