Advancing Web Accessibility with Machine Learning

No Thumbnail Available

URL

Journal Title

Journal ISSN

Volume Title

School of Arts, Design and Architecture | Bachelor's thesis
Ask about the availability of the thesis by sending email to the Aalto University Learning Centre oppimiskeskus@aalto.fi

Date

2024

Department

Major/Subject

Muotoilu

Mcode

Degree programme

Language

en

Pages

43

Series

Abstract

My thesis explores web accessibility and how machine learning can enhance it. As web accessibility becomes increasingly relevant, new ways to leverage machine learning are continually sought. This thesis focuses on a branch of machine learning: text classification models. As part of web accessibility involves websites' written content, their inaccessibility can significantly hinder users' ability to obtain information and engage with online re- sources. The thesis encompasses a comprehensive literature review regarding reading com- prehension, the factors influencing text readability, and the methods for measuring reada- bility. Additionally, the study delves into applying machine learning and text classification models to evaluate text based on its readability. The aim of this is to acquire the necessary knowledge to program a text classification model capable of identifying features that make text complex in terms of readability. This would provide proof of concept that a text classification model can be a valuable tool in making striving for accessible internet more effortless. The results underscore the potential of such a tool to aid web developers in evaluating the accessibility of their websites and to assist users by tailoring the readability of web content. The thesis culminates in a thorough discussion of the implications and limi- tations of the research, underscoring the pivotal role of machine learning in advancing web accessibility.

Description

Supervisor

Salovaara, Antti

Thesis advisor

Salovaara, Antti

Keywords

web accessibility, machine learning, text classification, accessibility, readability, text comprehension, Python

Other note

Citation