Search Results ams_dm_score_pct_results




Overview
Explore in AI Chat

Overview

The AMS_DM_SCORE_PCT_RESULTS table is a core data object within the Oracle E-Business Suite Marketing (AMS) module, specifically for releases 12.1.1 and 12.2.2. It functions as a repository for summarized statistical data generated by the Data Mining (DM) scoring engine. As its description indicates, it acts as a container for the percentile results of a scored population. This table is essential for analyzing the distribution and ranking of model scores across a target audience, enabling marketers to segment populations into meaningful groups, such as top deciles or quartiles, for targeted campaign execution.

Key Information Stored

The table's structure is designed to store aggregated percentile metrics for a given scoring run. The primary columns, as defined by its keys, are critical for understanding its data model. The SCORE_RESULT_ID serves as the unique primary key identifier for each percentile summary record. The SCORE_ID is a foreign key that links the percentile results back to a specific scoring model execution stored in the AMS_DM_SCORES_ALL_B table. The PERCENTILE column stores the specific percentile value (e.g., 90 for the 90th percentile) for which the summarized results are calculated. While the metadata excerpt does not list all columns, typical supporting data would likely include aggregated metrics like the count of records, average score, minimum score, and maximum score for each defined percentile within a specific SCORE_ID population.

Common Use Cases and Queries

The primary use case is post-scoring analysis to evaluate model performance and define actionable segments. A common operational query would retrieve the score thresholds for key percentiles to set campaign cutoffs. For example, to find the score defining the top 10% of a population for a specific model run, one might query: SELECT * FROM AMS_DM_SCORE_PCT_RESULTS WHERE SCORE_ID = :p_score_id AND PERCENTILE = 90;. For reporting, a join with the scores table allows analysts to generate distribution reports showing how a population is spread across percentiles, which is fundamental for lift analysis and campaign planning. These results drive decisions on which customer segments receive high-priority marketing communications.

Related Objects

The table maintains a strict dependency relationship with the central scoring table via a documented foreign key. The key related objects are:

  • AMS_DM_SCORES_ALL_B: This is the primary parent table. The relationship is defined as AMS_DM_SCORE_PCT_RESULTS.SCORE_ID references AMS_DM_SCORES_ALL_B. Every percentile summary record must be associated with a valid scoring model execution record in this table.
  • The table's integrity is enforced by its primary key constraint (AMS_DM_SCORE_PCT_RESULTS_PK on SCORE_RESULT_ID) and a unique key constraint (AMS_DM_SCORE_PCT_RESULTS_UK1 on SCORE_ID and PERCENTILE), ensuring no duplicate percentile summaries exist for a single scoring run.