Search Results ams_dm_model_type




Overview
Explore in AI Chat

Overview

APPS.AMS_DM_SCORE_HISTORY_V is a reporting and integration view within the Oracle E-Business Suite Oracle Marketing (AMS) module, specifically belonging to the Advanced Product Catalog and Marketing data-mining (AMS_DM) schema family. The view consolidates scoring history for parties that have been evaluated by Oracle Marketing data-mining models, joining source scoring records with their associated score definitions, model definitions, and lookup-based model type descriptions. Its principal function is to present a flattened, human-readable history of scores assigned to a party, enriched with the score name, the governing model name, and the meaning of the model type. The view is owned by the APPS schema, making it accessible to standard EBS application users and concurrent programs without requiring direct grants on the underlying AMS_DM base tables.

Underlying Base Objects

The view is defined over six documented objects, all referenced as APPS synonyms or views: AMS_DM_SOURCE, AMS_DM_MODELS_ALL_B, AMS_DM_MODELS_ALL_TL, AMS_DM_SCORES_ALL_B, AMS_DM_SCORES_ALL_TL, and AMS_LOOKUPS. AMS_DM_SOURCE is the driving table, supplying the party identifier, the continuous score, the decile, the percentile, and the audit columns, and it is filtered to ARC_USED_FOR_OBJECT = 'SCOR' so that only source rows tied to a score object are returned. The score identifier (USED_FOR_OBJECT_ID) is joined to AMS_DM_SCORES_ALL_B and its translation table AMS_DM_SCORES_ALL_TL, which supply the score name. The score's MODEL_ID is joined to AMS_DM_MODELS_ALL_B and AMS_DM_MODELS_ALL_TL to obtain the model name and the MODEL_TYPE code. Finally, AMS_LOOKUPS is joined on LOOKUP_TYPE = 'AMS_DM_MODEL_TYPE' to translate the model type code into a descriptive meaning.

Key Columns

The view exposes the following significant columns:

Common Use Cases and Queries

The view is typically queried to audit scoring outcomes, verify model behaviour, and feed downstream analytics. A frequent query pattern selects scoring history for a specific model type using the lookup joining logic embodied in the view:

SELECT party_id, score_name, model_name,
       model_type_code, model_type_meaning,
       continuous_score, decile, percentile
FROM   apps.ams_dm_score_history_v
WHERE  model_type_meaning = 'Predictive'
ORDER BY party_id, last_update_date DESC;

Because the view joins translation tables on USERENV('LANG'), results are returned in the session's current language, which makes it suitable for localized reporting. Analysts also use it to compare decile and percentile distributions across models, and to reconcile scores with their originating source records. Understanding the AMS_DM_MODEL_TYPE lookup is central, since the user's search term "ams_dm_model_type" corresponds directly to the lookup type that drives the MODEL_TYPE_MEANING column, enabling filtering and grouping by model category.