Affinity Profiling From Viewing Metadata for Real-Time Targeting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing TV content distribution systems face challenges in accurately determining user preferences and behaviors due to the diversity of content sources and platforms, leading to inefficient targeting and advertising, high churn rates, and the need for more sophisticated audience targeting capabilities.

Innovation Solution

A computer-implemented method generates affinity profiles for users based on their content selection and viewing habits using machine learning models, independent of demographic data, to determine intentions and preferences, enabling personalized content distribution and targeted advertising.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual analytical methods are used to determine user preferences and behaviors, then accuracy can be maintained, but processing time and operational costs increase significantly

Engineering Contradiction:

Improveaccuracy of user preference determinationVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual analytical methods with automated machine learning models and algorithms that process user data, content metadata, and viewing behaviors automatically. This substitution of mechanical human analysis with computational systems resolves the contradiction by maintaining measurement precision through sophisticated algorithms while dramatically reducing processing time and operational costs.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service through automated affinity profile generation where the machine learning models independently process and analyze user data without requiring manual intervention. The automated content recommendation engine serves itself by continuously learning from user interactions and generating recommendations, eliminating the need for manual analytical processes while maintaining high accuracy.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If diverse content sources and platforms are integrated to improve user understanding, then the comprehensiveness of user profiles increases, but system complexity and data processing challenges increase

Engineering Contradiction:

Improvecomprehensiveness of user profilesVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal content metadata schema and standardized data structures that can accommodate diverse content sources and platforms. The machine learning models are designed to process multiple data types and formats through a unified framework, enabling the system to integrate comprehensive user data from various sources while managing complexity through standardization and multi-functional processing capabilities.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Speed

If real-time processing of user actions is implemented to meet time constraints, then user experience improves, but computational resources and processing power requirements increase

Engineering Contradiction:

Improvereal-time processing speedVSAvoidcomputational resources

Core Design Contradiction:

SpeedVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by pre-processing and indexing content metadata, pre-computing affinity profiles based on available data, and preparing recommendation candidates in advance. This allows the system to deliver real-time recommendations by selecting and ranking pre-computed results rather than performing complete analysis at the moment of user interaction, thereby reducing computational resources while maintaining real-time performance.

Inventive Principle:

Principle #10Preliminary action

4Productivity

If automated machine learning models are deployed to generate affinity profiles, then productivity and scalability improve, but system complexity and implementation difficulty increase

Engineering Contradiction:

Improveprofile generation efficiencyVSAvoidimplementation complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces intermediary components such as feature engineering layers, data preprocessing pipelines, and model abstraction interfaces that simplify the deployment of machine learning models. These intermediaries act as mediators between raw data and complex algorithms, providing standardized interfaces and automated workflows that improve productivity while reducing implementation complexity through modular architecture and reusable components.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12610097B2Affinity profile system and method

Publication Date: 2026.04.21 THINKANALYTICS

AI summary

A computer-implemented method of determining affinity profiles, comprises:for each of a plurality of user devices, monitoring user activity including identifying content selected for viewing by the user of the user device;obtaining metadata concerning the selected items of content, the metadata representing at least some properties of the selected items of content;generating or updating a user record for the user, the user record comprising or representing the user activity and/or the associated content metadata;processing the user record to generate an affinity profile for the user based on the user record, the affinity profile comprising at least one selected affinity category or affinity category score, whereinthe affinity categories are selected from a stored set of affinity categories, each representing a user's affinity for a respective subject area.