AI-Powered Accessibility: Using Machine Learning to Detect and Correct Accessibility Gaps in Web Interfaces

Abstract

This research explores the integration of machine learning (ML) into the frontend development workflow to enhance web accessibility, ensuring compliance with standards like the Web Content Accessibility Guidelines (WCAG). We investigate ML-based techniques for detecting and correcting accessibility issues automatically, focusing on areas like alt-text generation, ARIA role enhancement, color contrast analysis, and adaptive interface testing. The paper also addresses technical challenges, ethical concerns, and future trends in AI-powered accessibility, presenting a framework for integrating intelligent accessibility scripts within development processes.