UI Accessibility Assessment Using Graph Embeddings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user interface (UI) accessibility guidelines lack actionable measurements for visual cues and design elements, making it difficult to create inclusive interfaces for users with disabilities, and manual assessment is error-prone and inefficient.
Innovation Solution
A UI assessment system using machine learning techniques to analyze UI elements, generate embeddings, and predict accessibility parameters, providing actionable feedback for interface improvements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual assessment of UI is used, then accessibility evaluation can be performed, but it is error-prone and inefficient when analyzing vast amounts of data
Solution Approach 1:
The patent replaces manual mechanical assessment with an automated machine learning system that uses computer vision and natural language processing to evaluate UI accessibility. The system automatically analyzes UI elements, generates accessibility reports, and provides recommendations without human intervention, thereby eliminating human error and significantly improving assessment efficiency while maintaining or enhancing measurement precision.
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a bridge between raw UI data and accessibility evaluation results. This intermediary system processes vast amounts of UI data, extracts relevant features, and transforms them into meaningful accessibility metrics, enabling both high precision and high productivity simultaneously.
2Ease of manufacture
If existing accessibility guidelines are used, then high-level structure recommendations are provided, but they cannot provide measurements for smaller scale changes such as positioning of UI elements
Solution Approach 1:
The patent segments the UI assessment process into multiple granular levels, analyzing individual UI elements (buttons, text fields, icons) separately rather than evaluating the overall UI structure only. This segmentation enables precise measurement of small-scale changes such as element positioning, sizing, and spacing while still providing high-level structural guidance, thereby resolving the contradiction between ease of implementation and measurement precision.
Solution Approach 2:
The patent adds a new dimension of analysis by evaluating UI elements across multiple spatial and functional dimensions simultaneously. The system analyzes not only the macro-level layout but also micro-level properties such as element positioning, font sizes, color contrast, and interactive areas, providing comprehensive measurements that existing guidelines cannot offer.
3Loss of information
If machine learning models are used to assess UI, then actionable feedback and pattern recognition are enabled, but system complexity increases
Solution Approach 1:
The patent employs a universal machine learning framework that performs multiple functions: it detects UI elements, analyzes accessibility compliance, identifies patterns across different UIs, generates recommendations, and provides actionable feedback. This multi-functional approach consolidates what would otherwise require multiple separate systems, thereby enabling comprehensive pattern recognition while managing overall system complexity.
4Reliability
If iterative manual assessment by multiple testers is performed, then comprehensive accessibility coverage is achieved, but time consumption and resource requirements increase
Solution Approach 1:
The patent implements continuous automated assessment that can evaluate UIs without interruption and process multiple UIs simultaneously. The machine learning system operates continuously, analyzing UI elements in real-time and providing immediate feedback, thereby achieving comprehensive accessibility coverage without the time delays inherent in iterative manual assessment processes.
Data Source
AI summary
Methods and systems are described herein for a user interface assessment (UI) system that assesses interfaces using machine learning models. The UI assessment system may receive a request for evaluating a UI comprising (1) a representation of the UI and (2) user-defined target values for parameters. The system may identify elements of the UI and structural properties of each element, each structural property controlling presentation of a corresponding element. The system may generate, for each element, (1) a corresponding composite embedding and (2) a corresponding content embedding. The system may generate a graph representation. The system may then generate a graph embedding using an embedding model. The graph embedding may be input into a parameter analysis machine learning model to obtain a corresponding predicted value for each of the parameters and transmit a corresponding predicted value for each of the parameters.