Dynamic Audience Segmentation Using Attribute Cardinality Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems face challenges in optimally segmenting large volumes of user data due to high cardinality attributes, leading to suboptimal decision-making in content targeting and recommendations, as conventional heuristics often lack data-driven support.

Innovation Solution

A dynamic segmentation system that analyzes attributes for cardinality, generates candidate attributes through supervised reduction and clustering, and allows user supervision to adjust parameters for optimal segmentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional heuristics are used for user segmentation, then the segmentation process is simple to implement, but the accuracy and effectiveness of audience grouping deteriorates due to lack of data-driven support

Engineering Contradiction:

Improveease of implementationVSAvoidsegmentation accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system implements feedback by continuously analyzing user interaction data and using the results to dynamically adjust and refine audience segments. The segmentation process incorporates feedback loops where segmentation performance is evaluated and used to improve future segmentation, moving from static heuristics to dynamic, data-driven adjustments

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system changes parameters by dynamically adjusting segmentation criteria based on analyzed user data patterns. Instead of fixed heuristic rules, the system modifies segmentation parameters (such as grouping criteria, attribute weights, and segment definitions) based on empirical data analysis, allowing the segmentation model to adapt to actual user behavior patterns

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If high cardinality attributes are used for segmentation, then the granularity and detail of user grouping is improved, but the complexity of the segmentation system increases

Engineering Contradiction:

Improvesegmentation granularityVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies segmentation by breaking down high-cardinality attributes into manageable groups or bins. Instead of treating each unique value separately, the system segments similar values into categories, reducing the effective cardinality while preserving meaningful distinctions for audience grouping

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces intermediary processing layers between raw high-cardinality attributes and final segmentation results. These intermediaries include data analysis components that aggregate, cluster, or transform high-cardinality attributes into lower-cardinality representations that are more suitable for practical segmentation while retaining essential information

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If dynamic data-driven segmentation is implemented, then the effectiveness of content targeting is improved, but the computational resources and processing time required increase

Engineering Contradiction:

Improvecontent targeting effectivenessVSAvoidcomputational resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by pre-analyzing and pre-segmenting user data before actual content delivery scenarios arise. User interaction data is processed and segmented in advance, creating ready-to-use audience groups that can be quickly applied during content targeting without requiring intensive real-time computation

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements dynamics by enabling flexible, adjustable segmentation models that can adapt to changing data patterns and requirements. The segmentation framework allows dynamic modification of segments based on new data while maintaining computational efficiency through incremental updates rather than complete re-processing

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12493633B2Dynamic cardinality-based group segmentation

Publication Date: 2025.12.09 TEALIUM

AI summary

Systems and methods are provided for analysis and selection of attributes used to segment data entities. The attributes used to segment data entities may be analyzed to identify segments of data entities (e.g., distinct audiences of visitors) that share values for a given subset of attributes. By intelligently selecting attributes for use in the segmentation process based on the values that they may take (e.g., the cardinality of the attributes), the selected attributes can be used to generate a reasonable or otherwise desirable number of data entity segments. Other attributes can be excluded from the segmentation process.