Recurring Retrieval Needs in Diverse and Dynamic Dataspaces: Issues and Reference Framework

Processing information requests over heterogeneous dataspaces is very challenging because aimed at guaranteeing user satisfaction with respect to conflicting requirements on result quality and response time. In [3], it has been argued that, in dynamic contexts preventing substantial user involvement in interpreting requests, information on similar requests recurring over time can be exploited during query processing. In this paper, referring to a graph-based modeling of dataspaces and requests, we propose a preliminary approach in this direction centered on the enabling concept of Profiled Graph Query Pattern (PGQP) as an aggregation of information on past evaluations of a set of previously executed queries. The information maintained in PGQP is not query results, as in materialized queries, but can include different kinds of data and metadata.