Skip to content

Entity Resolution

Definition

Asserts that records with similar values across the configured target fields are resolved as the same entity and share a single distinction field value.

Overview

Entity Resolution is a multi-field rule. You pick one or more target fields that describe the entity (for example, name, address, phone), choose how each field is compared (fuzzy text, numeric proximity, datetime tolerance, or exact (blocking)), and the platform clusters records whose weighted similarity meets the composite match threshold. Each cluster is assigned a unique cluster identifier (_qualytics_entity_id).

Once clusters are built, the rule checks the distinction field: every record in the same cluster must share the same value of the distinction field. Clusters that hold more than one value of the distinction field are flagged.

Typical use cases:

  • Match customer or company records with name and address variations.
  • Consolidate duplicate entities across systems that record slightly different spellings.
  • Identify fuzzy matches for deduplication before pushing records downstream.

Field Scope

Calculated: Entity Resolution does not take a fixed list of fields. Instead, the platform derives the evaluated fields from the target fields you configure (each entry names a single field plus a comparison strategy). The distinction field is configured separately.

Distinction Field: Accepted Types

Type Supported
Date
Timestamp
Integral
Fractional
String
Boolean

Target Field Types

Target Field Type How It's Compared
String fuzzy (default): text similarity, optionally promoted to a perfect match by substring containment or phonetic (homophone) match. Term-frequency weighting can also be enabled to reduce the impact of common tokens. exact: blocking pre-filter.
Numeric absolute (default): pair matches if the difference is within a fixed delta. relative: pair matches if the difference is within a percentage. exact: blocking pre-filter.
DateTime offset (default): pair matches if both timestamps are within a number of seconds. granularity: pair matches if both timestamps fall in the same Day, Week, Month, or Year. exact: blocking pre-filter.

General Properties

Name Supported
Filter
Allows the targeting of specific data based on conditions
Coverage Customization
Allows adjusting the percentage of records that must meet the rule's conditions

The filter allows you to define a subset of data upon which the rule will operate.

It requires a valid Spark SQL expression that determines the criteria rows in the DataFrame should meet. This means the expression specifies which rows the DataFrame should include based on those criteria. Since it's applied directly to the Spark DataFrame, traditional SQL constructs like WHERE clauses are not supported.

Examples

Direct Conditions

Simply specify the condition you want to be met.

Correct usage" collapsible="true 
O_TOTALPRICE > 1000

C_MKTSEGMENT = 'BUILDING'
Incorrect usage" collapsible="true 
WHERE O_TOTALPRICE > 1000

WHERE C_MKTSEGMENT = 'BUILDING'

Combining Conditions

Combine multiple conditions using logical operators like AND and OR.

Correct usage" collapsible="true 
O_ORDERPRIORITY = '1-URGENT' AND O_ORDERSTATUS = 'O'

(L_SHIPDATE = '1998-09-02' OR L_RECEIPTDATE = '1998-09-01') AND L_RETURNFLAG = 'R'
Incorrect usage" collapsible="true 
WHERE O_ORDERPRIORITY = '1-URGENT' AND O_ORDERSTATUS = 'O'

O_TOTALPRICE > 1000, O_ORDERSTATUS = 'O'

Utilizing Functions

Leverage Spark SQL functions to refine and enhance your conditions.

Correct usage" collapsible="true 
RIGHT(
    O_ORDERPRIORITY,
    LENGTH(O_ORDERPRIORITY) - INSTR('-', O_ORDERPRIORITY)
) = 'URGENT'

LEVENSHTEIN(C_NAME, 'Supplier#000000001') < 7
Incorrect usage" collapsible="true 
RIGHT(
    O_ORDERPRIORITY,
    LENGTH(O_ORDERPRIORITY) - CHARINDEX('-', O_ORDERPRIORITY)
) = 'URGENT'

EDITDISTANCE(C_NAME, 'Supplier#000000001') < 7

Using scan-time variables

To refer to the current dataframe being analyzed, use the reserved dynamic variable {{ _qualytics_self }}.

Correct usage" collapsible="true 
O_ORDERSTATUS IN (
    SELECT DISTINCT O_ORDERSTATUS
    FROM {{ _qualytics_self }}
    WHERE O_TOTALPRICE > 1000
)
Incorrect usage" collapsible="true 
O_ORDERSTATUS IN (
    SELECT DISTINCT O_ORDERSTATUS
    FROM ORDERS
    WHERE O_TOTALPRICE > 1000
)

While subqueries can be useful, their application within filters in our context has limitations. For example, directly referencing other containers or the broader target container in such subqueries is not supported. Attempting to do so will result in an error.

Important Note on {{ _qualytics_self }}

The {{ _qualytics_self }} keyword refers to the dataframe that's currently under examination. In the context of a full scan, this variable represents the entire target container. However, during incremental scans, it only reflects a subset of the target container, capturing just the incremental data. It's crucial to recognize that in such scenarios, using {{ _qualytics_self }} may not encompass all entries from the target container.

Anomaly Types

Type Supported
Record
Flag inconsistencies at the row level
Shape
Flag inconsistencies in the overall patterns and distributions of a field

Next Steps

  • How It Works

    Full semantics: clustering behavior, blocking vs. fuzzy fields, weighted composite score, threshold tuning, filter behavior, and how the anomaly is reported.

    How It Works

  • Examples

    Three production scenarios with sample data, source records, anomaly messages, and the clustering logic the platform applies.

    Examples

  • API

    Payload shape and field notes for creating an Entity Resolution check programmatically.

    API

  • FAQ

    Short answers to questions about target fields, threshold tuning, source records, and anomaly reporting.

    FAQ