Dataset for paper Towards accessible website design through artificial intelligence: A systematic literature review
Description
The accompanying dataset provides full transparency and reproducibility of the systematic literature review presented in the article Towards Accessible Website Design Through Artificial Intelligence. It consists of three components: The folder "01-Related Work search" documents the initial automated search process conducted in the Scopus and Web of Science databases. It includes the records retrieved based on the filters applied during the preliminary review of related work. The folder "02-Automatic search results" contains the subsequent search stages carried out across selected digital library databases (Scopus, Web of Science, IEEE Xplore, ACM Digital Library, SpringerLink, and ScienceDirect). It includes exported bibliographic records organized by search strings, as well as inclusion and exclusion steps applied. The file "Dataset_Towards_accessible_website_through_AI.xlsx" serves as the main dataset. It contains the fully curated data resulting from the entire review process, including metadata for the 31 selected primary studies, quality assessment scores, AI techniques employed, accessibility targets, and evaluation methods. It also includes the search queries, filters used, inclusion/exclusion criteria, and the bibliometric and synthesis analyses performed across the three phases of the proposed method: planning, conduction, and reporting. Together, these materials support the reproducibility of the study and provide a structured foundation for future meta-analyses or replication studies in the field of AI-driven web accessibility.