Sequential Graph Models for Device and Household Game Targeting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Simple rule-based advertisement or recommender systems fail to accurately capture the specific preferences of users regarding device and content interactions, particularly in gaming, due to challenges in recognizing game sessions and determining user preferences dynamically.

Innovation Solution

A machine learning-based approach using a sequential graph-based model to analyze device-level and household-level preferences, incorporating a scoring metric to filter trustworthy gaming exposures and generate game segments for targeted advertising.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If simple rule-based advertisement or recommender systems are used, then device complexity is reduced, but measurement precision of user preferences deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoiduser preference accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces rule-based mechanical systems with machine learning models that automatically learn user preferences from behavioral data. The sequential graph-based model and feature engineering pipeline substitute manual rule configuration with automated statistical learning, improving measurement precision while managing complexity through algorithmic approaches.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes parameters by transitioning from fixed rule-based thresholds to dynamic, data-driven preference scores. The machine learning models continuously adjust preference parameters based on observed user behavior, enabling adaptive measurement that improves accuracy without requiring proportional increases in system complexity.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If machine learning models with feature engineering pipelines are implemented, then measurement precision of gaming exposures is improved, but device complexity increases

Engineering Contradiction:

Improvegaming exposure accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex measurement task into distinct components: device-level features, household-level features, and sequential behavior patterns. This modular feature engineering approach allows each segment to be processed independently by specialized machine learning models, improving measurement precision while managing overall system complexity through division of labor.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces intermediary processing layers including feature engineering pipelines and sequential graph-based models that mediate between raw user behavior data and final preference measurements. These intermediaries transform complex raw data into structured features, improving measurement accuracy while isolating the complexity within manageable intermediate processing stages.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If multi-scale multi-granular targeting is implemented, then adaptability to user preferences is improved, but difficulty of detecting and measuring increases

Engineering Contradiction:

Improvetargeting flexibilityVSAvoidpreference detection complexity

Core Design Contradiction:

Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent adds dimensional layers to the targeting system by implementing multi-scale (individual user, household, device) and multi-granular (broad preferences, specific game genres, time-of-day patterns) analysis dimensions. This dimensional expansion enables comprehensive adaptability to diverse user preferences while systematically organizing the detection and measurement complexity across multiple hierarchical levels.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The machine learning framework implements universal models that can detect and measure various types of user preferences across different scales and granularities using the same underlying architecture. The sequential graph-based model and feature engineering pipeline serve multiple functions simultaneously, detecting device preferences, household patterns, and temporal behaviors within a unified system that reduces overall detection difficulty.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12415135B2Effective multi-scale multi-granular targeting for game users

Publication Date: 2025.09.16 SAMSUNG ELECTRONICS CO LTD

AI summary

A method includes obtaining, based on a sequential graph-based model, gaming exposure information over time, where the gaming exposure information includes device-level preferences and household-level preferences. The method also includes combining one or more raw user behavior sessions into a gameplay session based on the obtained gaming exposure information. The method further includes providing a scoring metric to (i) check an extent of multi-matching in the obtained gaming exposure information and (ii) remove untrustworthy gaming exposures from the obtained gaming exposure information. In addition, the method includes generating, based on a feature engineering pipeline, one or more game segments running in a production environment, where the one or more game segments are identified for ancillary content based on inferences by a machine learning model trained using the gaming exposure information.