Configure

Hover over the Configure tab on the navigation bar to see configuration options. The options are:

Match Configurations: Adjust the parameters that determine the match score between two tokens. Sets of parameter values can be saved as match configurations and applied to indices.
Overrides: Add or remove token pairs that are considered a match regardless of similarity.
Stopwords: Add or remove words that are ignored during string matching so that they do not affect the match score.

Match Configurations

Hover over the Configure tab and select Match Configurations to view the Match Configurations page.

Individual name tokens are scored by a number of algorithms. These algorithms can be optimized by modifying configuration parameters, thus changing the final match score. RMS provides an easy way to compare and test different parameter settings based on your data.

A match configuration contains a set of parameters. Each named match configuration contains parameter values for a specific language pair and entity type. A single named match configuration can contain multiple language pairs and entity types.

The Configure page contains a list of existing match configuration files. You can:

Edit an existing match configuration. You cannot edit a match configuration if an evaluation has already been created using the configuration. For the default match configuration, which cannot be edited, this option is replaced with View.
Copy an existing match configuration, then edit it, making the changes you'd like to test. This is how you modify an existing match configuration that has been used in an evaluation.
Export the match configuration as a .yaml file. This file can be re-imported into RMS or used as a reference for making a configuration in RNI.

Important

The exported .yaml file cannot be imported directly into RNI.
Delete a match configuration. This option is disabled for the default configuration.
Create a New Match Configuration.
Import Configuration from valid .yaml file.

Parameters

While RNI contains over 100 parameters, RMS exposes a few of the most commonly modified parameters that are the most likely to have an impact on your data.

RMS Field Name	Description	Behavior	RNI Parameter Name
Initials Score	The score that is assigned to an initial matching a token.	Increasing leads to higher final score when there is a match between an initial and a token.	`initialsScore`
Token Deletion Score	The score applied to an unmatched token when surrounding tokens are matched.	Increasing leads to higher final score when there is an unmatched token.	`deletionScore`
Token Reorder Penalty	A penalty applied when two tokens match, but are in different positions in each name.	Increasing leads to lower final score when the tokens aren't in the same position.	`reorderPenalty`
Token Conflict Score	The score that is assigned to unmatched conflict tokens.	Increasing leads to higher final score.	`conflictScore`
Right End Boost	A boost applied to the final score applied to the weights of tokens at the right end of the name.	Increasing leads to a higher score for matches between the surnames in English and other languages where the end of the name is more meaningful. It makes the last token more important.	`boostWeightAtRightEnd`
Initials Conflict Score	The score that overrides the usual conflict score when initials conflict.	Increasing leads to higher final score when initials conflict.	`initialsConflictScore`
Initialism Score	The score that gets assigned to an initialism matching a name.	Increasing leads to a higher score for matches between initialisms and names.	`initialismScore`
Stuck Initial Score	The score that gets applied when an initial is "stuck" to the preceding token.	Increasing leads to a higher score when there is a stuck initial.	`stuckInitialScore`
Out of Order Token Deletion Score	The score that gets assigned to an unmatched token which, when removed, leaves the remaining tokens out of order.	Increasing leads to a higher score when there is an out of order deletion.	`outOfOrderDeletionScore`
Initials Deletion Penalty	Adjusts the usual deletion score when initials are deleted by multiplying by this amount.	Increasing leads to a higher score when initials are deleted.	`initialsDeletionPenalty`
Gender Conflict Penalty	A penalty applied when the gender of the names doesn't match.	Increasing leads to a lower score when the genders of the names do not match.	`genderConflictPenalty`
Cross-Language Gender Conflict Penalty	A penalty applied for cross language name matching. This should usually be lower than the Gender Conflict Penalty because some first names appear as different genders in different languages.	Increasing leads to a lower score when the genders of the names do not match across languages.	`crossLanguageGenderConflictPenalty`
Both Ends Boost	A boost applied when the tokens at the beginning and end match. If set too high, middle names may not be matched correctly or may be ignored.	Increasing leads to a higher score when tokens at the beginning and end match.	`boostWeightAtBothEnds`
Missing Tokens Sensitivity	Multiplies token scores of deleted tokens by this amount if they only occur in only one of the names (but more than one token remains).	Increasing leads to higher scores for deleted tokens if they only occur in one name and more than one token remains.	`adjustOneSidedDeletionScores`
Token Reorder Correction	Adjustment made when the tokens in one of the names have been rotated with respect to the other (e.g. A B C D vs D A B C).	Increasing leads to a higher score when tokens have been rotated.	`reorderCorrection`
Final Score Bias	This is used to normalize the scores so that they have roughly the same value from one release to the next, as well as between language pairs.	Increasing leads to a higher final score.	`finalBias`

New Match Configuration

Select the entity type from the top of the window (Person, Organization, or Location).
Select a language pair from the drop-down.
For each parameter you can adjust the score, either by moving the slider or editing the value in the box.
Continue for other language pairs or entity types, as desired.
Save or Save As

Each configuration must have a name. If you are creating a new configuration, only Save As will be available. Once you've saved it, you can modify and save the configuration under the same name.

Import Match Configuration

You can create a new match configuration by importing a valid import file. A match configuration created in this way will have all parameters defined in the file, including parameters not exposed in the RMS UI.

Note

You must have a valid import file. A valid import file is in .yaml format and follows the structure of a RNI-ES parameter configuration file. You can create a valid import file by exporting the default configuration and editing it by adding or modifying parameters.

Select Configure from the navigation bar.
Select Import Configuration.
Name the match configuration.

Note

If the name of an existing configuration is provided and the configuration is editable, its values will be overwritten by the import. If the name of an existing configuration is provided and the configuration is not editable, the import will fail. Otherwise, a new match configuration will be created with values from the imported file.

A match configuration is not editable if an evaluation has already been created using the configuration.
Select OK. The new match configuration appears in the match configuration list.

Export Match Configuration

Exported match configurations can be re-imported into RMS or used as a reference for making a configuration in RNI.

Important

The exported .yaml file cannot be imported directly into RNI.

Select Configure from the navigation bar.
Select Export in the Options column for the desired match configuration.

Overrides

Hover over the Configure tab and select Overrides to view the Overrides page. This page allows you to add or remove token pairs that are considered a match regardless of similarity. You can use this list for things like names with unusual nicknames, such as "Margaret" and "Peggy."

To add a new override:

Select New Override.
Select the language pair. This is the language of each token in the pair. A pair can be two tokens in the same language.
Select the entity type. Supported entity types are PERSON, ORGANIZATION, and LOCATION.
Enter the token for the first language in the language pair.
Enter the token for the second language in the language pair.
Select Save & Close or Save & Add Another.

To find an override:

Select the tab for the override's entity type.
Select the language pair for the override using the dropdown menu.
Search for a token in the override pair using the Search token field. This search field will not show results outside the selected entity type and language pair.

To delete an override:

Find the override using the procedure above.
Select Delete in the Action column for that override.

Stopwords

Hover over the Configure tab and select Stopwords to view the Stopwords page. This page allows you to add or remove words that are ignored during string matching so that they do not affect the match score. You can use this list for parts of names or titles that are not significant when determining a match, such as "Mr." or "Prime Minister."

To add a new stopword:

Select New Stopword.
Enter the stopword in the Token field.
Select whether the stopword is a prefix or a regex.
- Prefix: A word, letter, or number that is placed before another. An example is "Mr."
- Regex: Regular expression. A string of characters that defines a search pattern. An example is "\bfnu\b."
Select the language of the stopword.
Select the stopword's entity type. Supported entity types are PERSON, ORGANIZATION, and LOCATION.
Select Save & Close or Save & Add Another.

To find a stopword:

Select the tab for the stopword's entity type.
Select the stopword's language using the dropdown menu.
Select whether the stopword is a prefix or regex using the radio buttons.
Search for the stopword using the Search token field. This search field will not show results outside the selected entity type, language pair, and prefix/regex selection.

To delete a stopword:

Find the stopword using the procedure above.
Select Delete in the Options column for that stopword.

Articles in this section

Match Configurations

Important

Parameters

New Match Configuration

Import Match Configuration

Note

Note

Export Match Configuration

Important

Overrides

Stopwords

Comments

Articles in this section

Match Configurations

Important

Parameters

New Match Configuration

Import Match Configuration

Note

Note

Export Match Configuration

Important

Overrides

Stopwords

Related articles