Deep-Learning Approaches for Optimized Web Accessibility: Correcting Violations and Enhancing User Experience

Calista Huang, Alyssa Ma, Suchir Vyasamudri, Eugenie Puype, Sayem Kamal, Salar Cheema, Michael Lutz

Primary Area: datasets and benchmarks

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: web accessibility, artificial intelligence, large language models, benchmark, GPT

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: Using LLMs and prompt engineering to automate website accessibility violation corrections for individuals with impairments

Abstract: With the increasing need for inclusive, user-friendly technology, web accessibility is crucial to ensuring equal access to online content for individuals with disabilities, including visual, auditory, cognitive, or motor impairments. Despite the existence of accessibility guidelines and standards such as Web Content Accessibility Guidelines (WCAG) and the Web Accessibility Initiative (W3C), over 90% of websites still fail to meet the necessary accessibility requirements. Manually detecting and correcting accessibility violations can be time-consuming and error-prone, highlighting the need for automated and intelligent solutions. While research has demonstrated methods to find and target accessibility errors, limited research has focused on effectively correcting accessibility violations. This paper presents an automatic deep-learning-based approach to correcting accessibility violations in web content. We aim to enhance web accessibility, promote inclusivity, and improve the overall user experience for individuals with impairments. We employ website accessibility violation data and prompt engineering to identify potential accessibility issues within HTML code. Leveraging accessibility error information, large language models (LLMs), and prompt engineering techniques, we achieved an over 50% reduction in accessibility violation errors after corrections. While our research successfully illustrates the ability of prompt engineering techniques to efficiently correct website accessibility violation errors, further research may be necessary to explore a larger range of website URLs or to focus on researching techniques for best handling specific common accessibility errors. Our work demonstrates a valuable approach toward the direction of inclusive web content, and provides directions for future research to explore advanced methods to automate web accessibility.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 8875