Accessible Interface Content Generation for Personalized UI Adaptation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional accessibility guidelines, such as WCAG, fail to account for individual user preferences and are inadequate for non-web platforms, leading to inaccessible digital content and potential legal and reputational risks for organizations.

Innovation Solution

A system utilizing machine learning techniques to generate tailored, accessible interface content by determining user populations and platform preferences, modifying interface components to enhance accessibility across various platforms, including web, mobile, and desktop applications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional WCAG guidelines are used to ensure accessibility, then compliance with accessibility standards is achieved, but the system fails to account for individual user preferences and varying levels of disabilities

Engineering Contradiction:

Improveaccessibility complianceVSAvoidindividual user customization

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic accessibility adjustments by using machine learning models to continuously adapt interface content based on individual user preferences and interactions. The system transitions from static WCAG compliance to dynamic customization where interface features such as text size, contrast, and layout are automatically adjusted in real-time based on user-specific needs identified through ML analysis.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent applies local quality by customizing specific interface content components rather than applying uniform accessibility adjustments across the entire interface. The ML framework identifies and modifies only those interface elements that require adjustment for individual users, such as modifying text properties for visually impaired users while leaving other elements unchanged, thereby achieving personalized accessibility without unnecessary modifications.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If WCAG guidelines are applied to web content, then web accessibility is improved, but the system fails to address other technology platforms such as native mobile applications or desktop applications

Engineering Contradiction:

Improveplatform coverageVSAvoidaccessibility effectiveness

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent implements universality by creating a platform-agnostic ML framework that can generate and adapt interface content across multiple technology platforms including web, mobile, and desktop applications. The system uses unified accessibility models that can be deployed across different platforms, ensuring consistent accessibility effectiveness regardless of the specific platform being used.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If tailored interface content is generated using machine learning, then individual user accessibility needs are addressed, but the system complexity increases

Engineering Contradiction:

Improveuser personalizationVSAvoidsystem architecture

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary ML framework that sits between the user and the interface content generation process. This intermediary system handles the complexity of analyzing user preferences, determining accessibility needs, and generating customized interface content, thereby shielding end users from the underlying system complexity while still delivering personalized accessibility.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements self-service by enabling the ML system to automatically analyze user interactions, determine accessibility preferences, and generate customized interface content without requiring manual configuration or intervention. The system learns and adapts to individual user needs autonomously, reducing the operational complexity burden on users and administrators.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS12572730B2Systems and methods for generating accessible interface content using a machine learning framework

Publication Date: 2026.03.10 WELLS FARGO BANK NA

AI summary

Systems, apparatuses, methods, and computer program products are disclosed for providing accessible interface content. An example method includes receiving base interface content comprising one or more interface content components and determining a user population of interest and a platform of interest. The example method further includes generating one or more interface feature sets using a pre-processing model. The example method further includes modifying one or more interface content components for each interface feature set using feature modification models. The example method further includes generating modified interface content using a multimodal model based on the one or more interface feature sets and providing the modified interface content.