Results for “upper_boundary”

AI-generated from documented ETRM metadata — verify critical details on the linked pages.

Overview

The AMS_DM_NUMERICAL_BB table resides in the AMS (Marketing) schema of Oracle E-Business Suite and stores bin boundary definitions for numerical columns used within data mining data sources. In the Oracle Marketing module, data mining enables marketers to segment customers based on predictive models and analytical scoring. Numerical columns feeding these models often require discretization — the process of converting continuous numeric ranges into discrete bins or buckets — so that classification algorithms can process them. This table records the boundary values (lower and upper) that define each bin for a given numerical column, along with the bin identifier that ties the boundary to a specific bin definition.

From a Data Vault modeling perspective, the heuristic classification derived from the foreign key structure identifies this table as standalone. It references FND_SECURITY_GROUPS for security scoping, but is not itself a hub, link, or satellite in a strict Data Vault sense. Modelers integrating this data into an enterprise warehouse should treat it as a reference or lookup table rather than a transactional fact, recognizing that the security group reference is the sole documented outbound relationship.

Key Information Stored

The table contains 13 documented columns. The surrogate primary key is NUMERICAL_BB_ID, enforced by the constraint AMS_DM_NUMERICAL_BB_PK. A unique index, AMS_DM_NUMERICAL_BB_U1, also exists on NUMERICAL_BB_ID, reinforcing its role as the primary business-key candidate.

• NUMERICAL_BB_ID — Surrogate identifier and primary key for each bin boundary record.
• COLUMN_NAME — The name of the numerical column in the data mining data source to which these bin boundaries apply.
• LOWER_BOUNDARY — The lower numeric value defining the start of the bin interval.
• UPPER_BOUNDARY — The upper numeric value defining the end of the bin interval.
• BIN_ID — Identifier linking the boundary range to a specific bin definition.
• IS_ATTRIBUTE — Flag indicating whether the associated column is treated as an attribute in the mining model.
• SECURITY_GROUP_ID — Foreign key to FND_SECURITY_GROUPS, controlling multi-org or security-based access to the record.
• OBJECT_VERSION_NUMBER — Optimistic locking column used by the Oracle Applications framework to detect concurrent updates.
• LAST_UPDATE_DATE, LAST_UPDATED_BY, CREATION_DATE, CREATED_BY, LAST_UPDATE_LOGIN — Standard EBS audit columns tracking who created and last modified the record and when.

Common Use Cases and Queries

Typical use cases involve reviewing, validating, or reporting on how numerical attributes are bin-discretized for data mining models. Analysts may need to confirm the exact bucket ranges applied to a column such as customer age, purchase frequency, or monetary value before interpreting model outputs. A representative query retrieving all boundaries for a given column is:

• SELECT column_name, bin_id, lower_boundary, upper_boundary FROM ams.ams_dm_numerical_bb WHERE column_name = :col ORDER BY bin_id;
• Joining to FND_SECURITY_GROUPS to enforce organizational security: ... FROM ams.ams_dm_numerical_bb bb, fnd_security_groups sg WHERE bb.security_group_id = sg.security_group_id;
• Audit and change-tracking reports leveraging LAST_UPDATE_DATE and LAST_UPDATED_BY to identify recently modified bin definitions.
• Data quality checks confirming that boundary ranges do not overlap and cover the intended domain for each mining column.

Related Objects

The FK structure documents a single direct reference, but the table participates in the broader AMS data mining object family. The most significant related objects include:

• FND_SECURITY_GROUPS — referenced via SECURITY_GROUP_ID; governs security scoping of bin boundary records.
• Sibling AMS data mining definition tables (such as numerical column and data source definition tables in the AMS_DM_* family) that share the same mining source context.
• The AMS Marketing data mining engine and related PL/SQL APIs that read bin boundary metadata during model execution and scoring.
• Concurrent programs and reports within Oracle Marketing that generate or consume discretized numerical attributes.

Because the metadata classifies the object as standalone, direct dependencies outside the documented security group reference should be verified against the specific implementation before being relied upon in integration or migration work.