Investigation of Creating Accessibility Linked Data Based on Publicly Available Accessibility Datasets

With the fast growth of web, the web is full of diverse data. Linked Data is data on the web that gives URIs to entities and links different data from different domains together, which results in the part of the web infrastructure. Since the dramatic development of Linked Data movement, a great number of accessibility datasets have been published as open data on the web by governments or organizations so that they can be accessed and reused by anyone without any restriction. However, the majority of the accessibility datasets available on the web are in diverse formats such as HTML table, Excel and CSV, rather than Linked Data format – RDF. This paper mainly investigates the method of creating accessibility Linked Data based on the publicly existing available accessibility datasets. First of all, requested accessibility data was extracted from the accessibility dataset. Moreover, Jena RDF API is applied to create RDF documents based on the extracted data from the accessibility dataset. Compared to other approaches such as transforming XML to RDF via schema, current method of this project does not have mapping work and it shows significant convenience. Finally, a triple store was set up to sever the RDF documents. In addition, the server has query interface support for test; as a result, users could query the data they need. In the future development, relevant dataset links would be added to current RDF documents. At last the accessibility RDF dump would be published on the web as a part of the web infrastructure. Index Terms—Linked Data, Open Data, Semantic Web, Accessibility Dataset.