Data quality evaluation through data quality rules and data provenance.

The application and exploitation of large amounts of data play an ever-increasing role in today’s research, government, and economy. Data understanding and decision making heavily rely on high quality data; therefore, in many different contexts, it is important to assess the quality of a dataset in order to determine if it is suitable to be used for a specific purpose. Moreover, as the access to and the exchange of datasets have become easier and more frequent, and as scientists increasingly use the World Wide Web to share scientific data, there is a growing need to know the provenance of a dataset (i.e., information about the processes and data sources that lead to its creation) in order to evaluate its trustworthiness. In this work, data quality rules and data provenance are used to evaluate the quality of datasets. Concerning the first topic, the applied solution consists in the identification of types of data constraints that can be useful as data quality rules and in the development of a software tool to evaluate a dataset on the basis of a set of rules expressed in the XML markup language. We selected some of the data constraints and dependencies already considered in the data quality field, but we also used order dependencies and existence constraints as quality rules. In addition, we developed some algorithms to discover the types of dependencies used in the tool. To deal with the provenance of data, the Open Provenance Model (OPM) was adopted, an experimental query language for querying OPM graphs stored in a relational database was implemented, and an approach to design OPM graphs was proposed.