Matching options

Options in this dialog box configure matching settings for an entity and its fields.

This dialog box opens when you click the Matching options button next to an entity in the Match and Group step settings.

Entity Fields shows the entity that you are configuring and fields within the entity. Click the entity or any of its field entries to configure match settings. You can add or remove entity fields for matching.

Entity Settings

Settings shown here may be configured while an entity is selected in the Entity Fields box. Settings configure an entity match rule. Match Rules mapped to multiple entities are combined by an AND logical operator.

Matching Method: Determines the criteria for considering entities in two records to match.

• Based on threshold: Entities are considered to match if the threshold value is equaled or exceeded.
• All True: Entities are considered to match if all algorithms are true.
• Any True: Entities are considered a match if any algorithm is true.

Missing Data: Specifies how blank fields affect whether the two fields being compared are considered matches.

• Ignore Blanks: Ignores blank fields.
• Compare Blanks: Counts a match if a field is blank in both records being compared. When one field but not the other is blank, then fields are not considered to match.
• Count As 100: A field matches if it is blank.
• Count As 0: The field does not match if it is blank.

Scoring Method: Determines how the matching algorithms match scores are calculated to come up with one match score for the field. Each matching algorithm is scored based on how closely it matches the two fields. Only standard matching rules use the scoring method.

• Average: Simple average give the same weight to all scores.
• Maximum: Uses the highest entity score.
• Minimum: Uses the lowest entity score.
• Weighted Average: Calculates the average based on the relative importance of scores being averaged. This smooths out scoring fluctuations.
• Vector Summation: Uses the vector summation of each child score to determine the score. The formula for calculation is: sqrt(a^2+b^2+c^2) / sqrt(n), where a, b, and c are the scores of three children, and n is the number of children.

Field Settings

Missing Data

• Ignore Blanks: Ignores blank fields.
• Compare Blanks: Includes blank fields. Counts a match if a field is blank in both records being compared.
• Count as 100: Counts a blank field as an exact match.
• Count as 0: Count a blank field as not a match.

Threshold Score: Specifies the minimum match score needed for the field to be considered a match. The field is given a match score based on how closely it matches the same field in an existing record.

Scoring Method: Determines how the matching algorithms match scores are calculated to come up with one match score for the field. Each matching algorithm is scored based on how closely it matches the two fields. Only standard matching rules use the scoring method.

• Average: Uses the average match score.
• Maximum: Uses the highest match score.
• Minimum: Uses the lowest match score.
• Vector Summation: Uses vector summation of the score of each algorithm to determine the match score. This scoring method is useful if you want a higher vector summation match score in one or more algorithms to get proportionately represented in the final match score. The formula for calculating the final score is: sqrt(a^2+b^2+c^2) / sqrt(n), where a, b, and c are the scores of three different algorithms, and n is the number of algorithms used.
• Weighted Average: Uses the weight of each matching method to determine the match score.

Select Algorithm: Match algorithms determine how a specific field in a record is compared to the same field in another record. The matching algorithms look for strings that exactly match a pattern. If you are using international data, we recommend using the exact matching algorithm with your matching rules or an algorithm specific for the language (such as Spanish Metaphone for Spanish language strings). You can use the exact matching method for almost any field, including custom fields.

The matching algorithm defines the logic that determines whether two fields match. Each matching algorithm is scored based on how closely it matches the two fields. For example, if you select exact matching and the two fields match, the match score is 100. You can select one or more algorithms to match fields. Selecting more than one algorithm provide for each algorithm against weaknesses of other algorithms. The defaults includes recommended selections for an identified field type.

Matching Method-to-Scoring Method Matrix

The table below shows the logical relationship between Matching Method and Scoring Method and how each combination changes the logic used during match processing.