14:00, Room N107 (Parc Club)
The representation of imperfection has always been a major issue in the field of database management. It is in fact well recognized that if we cannot represent imperfection, such as missing, imprecise and uncertain values, we risk to lose much valuable information. Object of the talk will be the management of uncertain data, and more specifically the topic of data integration with uncertainty. Uncertainty is a state of limited knowledge, where we do not know which of two or more alternative statements is true – for instance, we may not know with certainty the value of a data item when two heterogeneous databases show different values for it. Data integration is the process of providing the user with a unified view of data residing at different sources. While traditional data integration methods more or less explicitly consider uncertainty as a problem, as something to be avoided, some recent approaches treat uncertainty as an additional source of information, sometimes that is precious and that should be preserved. In this talk I will focus on the status of research on uncertainty management in data integration, presenting some recent results.