Machine Learning Framework for Cross-Platform Interface Accessibility

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional accessibility guidelines, such as WCAG, fail to account for individual user preferences and are not applicable to technology platforms beyond web content, leading to potential legal and reputational risks for non-compliance.

Innovation Solution

A machine learning-based evaluation model framework that considers user population preferences and evaluates interface content across various platforms, including web, native mobile applications, and desktop applications, using rendering, user population, and scoring models to identify accessibility issues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional WCAG standards are used to evaluate accessibility, then compliance with established guidelines is achieved, but individual user preferences and diverse user population needs are not considered

Engineering Contradiction:

Improvecompliance with accessibility standardsVSAvoidconsideration of individual user preferences

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent segments the evaluation process into multiple specialized models: a rendering model that processes interface content, a user population model that simulates diverse user experiences, and a scoring model that evaluates accessibility. This segmentation allows each model to specialize in specific aspects of accessibility evaluation, enabling both compliance checking and personalized user preference consideration simultaneously.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The user population model acts as an intermediary between the interface content and the evaluation criteria. It simulates how different user populations with varying disabilities and preferences experience the interface, translating raw interface features into user-specific accessibility assessments. This intermediary enables the system to consider individual preferences while maintaining overall compliance standards.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Stability of the object's composition

If WCAG guidelines are applied uniformly across all platforms, then consistency in evaluation is maintained, but platform-specific accessibility issues are not addressed

Engineering Contradiction:

Improveconsistency of evaluation standardsVSAvoidplatform-specific accessibility evaluation

Core Design Contradiction:

Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The rendering model dynamically adapts the evaluation process to different technology platforms (web, mobile, desktop) while maintaining core accessibility principles. The model adjusts its analysis based on platform-specific characteristics such as touch interfaces for mobile or keyboard navigation for desktop, enabling consistent accessibility evaluation across diverse platforms without requiring separate rigid guidelines for each.

Inventive Principle:

Principle #15Dynamics

3Measurement precision

If detailed evaluation of interface content for multiple user populations is performed, then comprehensive accessibility assessment is achieved, but computational complexity and processing time increase

Engineering Contradiction:

Improveaccessibility assessment accuracyVSAvoidevaluation model complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by pre-processing interface content through the rendering model to extract relevant accessibility features before user population-specific evaluation. This preliminary processing organizes and structures the interface data in advance, making subsequent multi-population evaluation more efficient and manageable, thereby reducing overall computational complexity while maintaining comprehensive assessment accuracy.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250299202A1Systems and methods for evaluating interface content using a machine learning framework

Publication Date: 2025.09.25 WELLS FARGO BANK NA

AI summary

Systems, apparatuses, methods, and computer program products are disclosed for evaluating interface content for a user population. An example method includes receiving the interface content comprising one or more interface content components. The example method further include determining a user population of interest and selecting an evaluation model framework based on the user population of interest. The example method further includes determining an accessibility score for the interface content based on the one or more interface content components using the evaluation model framework and determining whether the accessibility score satisfies an accessibility score threshold. The example method further includes providing an interface content evaluation report.