Quality-Driven Query Processing over Federated RDF Data Sources

The integration of data from heterogeneous sources is a common task in various domains to enable data-driven applications. Data sources may range from publicly available sources to sources within data lakes of companies. The added value generated by integrating and analyzing the data greatly depends on the quality of the underlying data. As a result, querying heterogeneous data sources as a way of integrating data enabling such applications needs to consider quality aspects. Quality-driven query processing over RDF data sources aims to study approaches which consider data quality description of the data sources to determine optimal query plans. In contrast to most federated query approaches, in quality-driven query processing the quality of an optimal plan and thus of the retrieved data, not only depends on efficiency typically measured as execution time but also on other quality criteria. In this work, we present the challenges associated with considering multiple quality criteria in federated query processing and derive our problem statement accordingly. We present our research questions to address the problem and the associated hypotheses. Finally, we outline our approach including an evaluation plan and provide preliminary results.