Using SPARQL and SPIN for Data Quality Management on the Semantic Web

The quality of data is a key factor that determines the performance of information systems, in particular with regard (1) to the amount of exceptions in the execution of business processes and (2) to the quality of decisions based on the output of the respective information system. Recently, the Semantic Web and Linked Data activities have started to provide substantial data resources that may be used for real business operations. Hence, it will soon be critical to manage the quality of such data. Unfortunately, we can observe a wide range of data quality problems in Semantic Web data. In this paper, we (1) evaluate how the state of the art in data quality research fits the characteristics of the Web of Data, (2) describe how the SPARQL query language and the SPARQL Inferencing Notation (SPIN) can be utilized to identify data quality problems in Semantic Web data automatically and this within the Semantic Web technology stack, and (3) evaluate our approach.