Analyzing online schema extraction approaches for linked data knowledge bases

In order to help data curators, data scientists, and other users in the domain of Linked Data to identify potentially new data sources, it is important to understand the corresponding data schema. The schema can help to determine the domain of the data (e.g. bibliographic data, geospatial data, etc.), its structure and more. We analyzed several strategies and systems which extract schemas in an online approach. Established systems using SPARQL endpoints for online schema extraction are limited when knowledge bases are very large or complex. Due to the growth of Linked Data, the knowledge bases has become larger and their structure more complex. Therefore, this paper will discuss some limitations of current strategies.