Personalized Ad Ranking via User Preference Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional Internet advertising models fail to provide personalized ad ranking to users, relying on aggregate user behavior and not adapting to individual user preferences, resulting in a non-specific ad service experience.

Innovation Solution

A method that receives user session information, modifies ad quality measures using correction factors based on user preferences, and ranks ads accordingly to provide personalized ad delivery, improving user experience and ad selection likelihood.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional aggregate-based ad targeting is used, then ad delivery simplicity is maintained, but ad personalization and user experience quality deteriorate

Engineering Contradiction:

Improvead personalizationVSAvoidad ranking system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the aggregate user behavior data into individual user profiles by creating separate ad selection behavior records for each user. This segmentation enables personalized ad ranking while maintaining the overall system structure, resolving the contradiction between personalization and complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces correction factors as an intermediary mechanism that adjusts ad quality measures based on individual user behavior. These correction factors act as a mediator between the simple aggregate targeting system and the desired personalized experience, enabling adaptation without requiring complete system redesign.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If predetermined ad ranking is used for all users, then system operation simplicity is maintained, but ad service quality and user engagement deteriorate

Engineering Contradiction:

Improvead ranking operationVSAvoidad service quality

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent performs preliminary analysis of user ad selection behavior before ad ranking to generate correction factors. This preliminary action enables the system to automatically adjust rankings based on user preferences without requiring complex real-time operations, maintaining ease of operation while improving service quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism where user ad selection behavior is continuously monitored and used to generate correction factors that adjust ad quality measures. This feedback loop improves ad service quality by adapting to user preferences while keeping the operational process straightforward through automated adjustments.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If aggregate user behavior data is used, then data collection simplicity is maintained, but user-specific ad targeting precision deteriorates

Engineering Contradiction:

Improvead targeting precisionVSAvoiduser-specific behavior information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent applies local quality by creating user-specific correction factors from individual ad selection behavior data. Instead of treating all users uniformly, the system tailors the ad quality measures to each user's specific preferences and patterns, thereby improving targeting precision while systematically capturing user-specific information.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS10445768B1User-based ad ranking

Publication Date: 2019.10.15 GOOGLE LLC

AI summary

Advertisement quality measures (e.g., predicted click through rates) are modified in accordance with a user's preferences with respect to domains to which the advertisements relate.