Tailoring Data Quality Models Using Social Network Preferences

To succeed in their tasks, users need to manage data with the most adequate quality levels possible according to specific data quality models. Typically, data quality assessment consists of calculating a synthesizing value by means of a weighted average of values and weights associated with each data quality dimension of the data quality model. We shall study not only the overall perception of the level of importance for the set of users carrying out similar tasks, but also the different issues that can influence the selection of the data quality dimensions for the model. The core contribution of this paper is a framework for representing and managing data quality models using social networks. The framework includes a proposal for a data model for social networks centered on data quality (DQSN), and an extensible set of operators based on soft-computing theories for corresponding operations.