Results for “source_id”

50+ results




Overview
Explore in AI Chat

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

Overview

AMS_DM_APPLY_STG is a staging table owned by the AMS schema within the Oracle E-Business Suite Marketing (AMS) product family. Its documented purpose is to hold scoring run results produced by the data mining engine during marketing campaign execution. In Oracle EBS 12.1.1 and 12.2.2, the Marketing module leverages Oracle Advanced Data Mining (ODM) components to score customer and prospect records against predictive models, and AMS_DM_APPLY_STG acts as the transient landing area for those scored outputs before downstream processing, list generation, or promotion into persistent marketing list tables.

From a heuristic Data Vault modeling perspective, the mined relationship data classifies this object as standalone. This suggests it functions as an independent staging artifact rather than a hub, link, or satellite participating in a modeled business key relationship chain. The staging pattern is consistent with an extract-load area where row-level scoring results are isolated before validation and assignment.

Key Information Stored

The documented physical schema in ETRM 12.2.2 exposes three columns, and a single unique constraint — AMS_DM_APPLY_STG_UK1 — enforced on the column SOURCE_ID. The documented columns are:

  • SOURCE_ID — The surrogate identifier for the staging row and the column underpinning the primary/unique key constraint AMS_DM_APPLY_STG_UK1. It identifies the source entity being scored (typically the customer, prospect, or contact record passed into the data mining apply operation). Because it is documented as a unique key, it is the primary business-key candidate for this table.
  • SCORE — The numeric score returned by the data mining engine for the associated source record. This is the predicted value produced by applying the model.
  • PROBABILITY — The probability value accompanying the score, indicating the likelihood associated with the predicted outcome produced by the scoring run.

Because the metadata documents only these three columns, the surrogate-versus-business-key distinction is narrow: SOURCE_ID serves as the documented unique identifier, while SCORE and PROBABILITY are purely descriptive result payload columns. There is no separate sequence-generated surrogate key distinct from SOURCE_ID in the documented schema.

Common Use Cases and Queries

The most frequent use is inspecting the freshly staged scoring output immediately after a data mining apply job completes, verifying score distributions, and promoting qualifying records into marketing lists. A representative query retrieves top-scoring source records:

  • SELECT SOURCE_ID, SCORE, PROBABILITY FROM AMS.AMS_DM_APPLY_STG ORDER BY SCORE DESC;
  • SELECT SOURCE_ID, SCORE, PROBABILITY FROM AMS.AMS_DM_APPLY_STG WHERE PROBABILITY >= :threshold;
  • SELECT COUNT(*) FROM AMS.AMS_DM_APPLY_STG; to confirm the apply run populated rows.

Typical reporting scenarios include campaign response-propensity ranking, segment sizing at a given probability cutoff, and troubleshooting scoring jobs that return unexpected or null scores. Because the table is a staging object, joins back to customer or contact master data are performed on SOURCE_ID against the upstream source table that fed the apply process.

Related Objects

The ETRM metadata documents AMS_DM_APPLY_STG as standalone with no foreign keys, so related objects are inferred by function rather than by enforced constraints:

  • Customer and prospect master tables in the Trading Community Architecture (HZ) schema, joined on SOURCE_ID, supplying the population being scored.
  • AMS Marketing list and campaign member tables, which consume scored output to build targeted lists.
  • AMS data mining model and model-apply definition objects that parameterize the scoring run.
  • Concurrent program and request log objects that record the data mining apply execution populating this staging table.

Because the object is a staging table, rows are typically transient; they are archived or purged following promotion of the scored results into the persistent marketing structures.