Machine-Learned Database Interaction Model for Targeted Content Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Centralized database systems struggle to effectively identify which entities are most likely to interact with presented content items, limiting the system's ability to provide targeted content recommendations.

Innovation Solution

A central database system trains a machine-learned model using entity and content item characteristics, along with interaction data, to identify a subset of entities most likely to interact with specific content items, and displays these items with interface elements that facilitate interactions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If centralized database systems present content items to entities, then content coverage is improved, but the ability to identify likely interactors deteriorates

Engineering Contradiction:

Improvecontent coverageVSAvoidinteraction prediction accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The system performs preliminary training of the machine-learned model using historical interaction data before presenting content items. The model is pre-configured with entity characteristics, content item characteristics, and interaction patterns, enabling accurate prediction to be made at the time of content presentation without requiring real-time analysis of all entities.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The machine-learned model acts as an intermediary between the content items and the entities. Instead of directly analyzing all entities against all content items, the model mediates by processing entity characteristics and content characteristics through learned patterns to produce interaction probability predictions, significantly improving measurement precision.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If the system analyzes all entity characteristics to predict interactions, then prediction accuracy is improved, but processing time deteriorates

Engineering Contradiction:

Improveinteraction prediction accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-processes and stores entity characteristics and content item characteristics in a structured format before content presentation. Historical interaction data is used to train the model in advance, so that during actual content presentation, the system only needs to input the pre-processed characteristics into the trained model rather than analyzing raw data in real-time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates a simplified representation of complex entity and content characteristics that can be efficiently processed by the machine-learned model. The model learns to map these characteristic representations to interaction probabilities, allowing fast prediction without requiring complex real-time analysis of all original data.

Inventive Principle:

Principle #26Copying

3Adaptability or versatility

If the system presents content items to all entities, then content distribution is improved, but engagement quality deteriorates

Engineering Contradiction:

Improvecontent distributionVSAvoidengagement quality

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system applies different content presentation strategies to different entities based on their characteristics and predicted interaction probabilities. Instead of uniform distribution, entities with higher predicted engagement receive content presentation while those with lower probabilities do not, improving engagement quality through localized optimization.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes the parameter of content distribution from uniform to probability-based selection. The machine-learned model outputs interaction probabilities for different entities, and the system uses these probability values to determine which entities should receive content items, thereby improving engagement quality through data-driven parameter adjustment.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250371435A1Machine-learned database interaction model

Publication Date: 2025.12.04 GUSTO INC

AI summary

A central database system trains a machine-learned model based on training data identifying entity characteristics of account holder entities, content item characteristics of a content item presented to the account holder entities, and interactions between the account holder entities and the presented content item. The central database system then identifies a target set of account holder entities, and applies the trained machine-learned model to the entity characteristics of each account holder entity of the target set of account holder entities, the entity characteristics of each of the account holder entities that previously interacted with the content item, and the content item characteristics of the content item to identify a subset of the target set of account holder entities for presentation of the content item. The content item is then displayed to the subset, the content item includes an interface element that, when selected, causes an interaction to take place.