Masterarbeit
Exploring Artificial Intelligence for Increased Accessibility of Dynamic or User-Generated Content
Completion
2027/12
Research Area
Students
Amirali Shaban Khamseh
Advisers
Lucas Schröder
Dr.-Ing. Sebastian Heil
Description
The rapid growth of the digital landscape has made managing large volumes of heterogeneous and multilingual content increasingly challenging for web applications. Conventional search engines, which often rely on literal string matching, have limited abilities to interpret semantic intent, accommodate synonyms, or handle multilingual content. Furthermore, standard search interfaces often fail to comply with accessibility guidelines or provide essential assistive features. This leads to significant barriers for both users with visual or cognitive disabilities and for users with limited multilingual proficiency.
This thesis aims to address these limitations by designing an AI-enabled pipeline for advanced information processing, enrichment, and information retrieval that supports a more inclusive discovery process. This requires adequate handling of heterogeneous content while maintaining precise search capabilities and providing suitable presentation of search results in an accessible way, for example, by including automatic translation or making use of plain language.
The objective of this thesis is the creation of a solution or combination of existing approaches to solve the problem of accessible, multilingual search interfaces in web applications. This comprises an analysis of the state of the art focusing on AI techniques for information processing, information retrieval, and accessible presentation, the demonstration of the feasibility of the solution through a prototypical implementation, and a suitable evaluation based on the usability and accessibility of this prototype and its compliance with requirements extracted through the literature research.
