Using web-based personalization on spatial data warehouses

Abstract

Spatial data warehouses (SDW) rely on extended multidimensional (MD) models in order to provide decision makers with appropriate structures to intuitively analyse spatial data. Several SDW development approaches provide a conceptual modeling and some guidelines in order to obtain logical schemas. However, there are two main drawbacks (i) spatial modeling is still complex for providing each decision maker with their own information needs, and (ii) SDW may be potentially large and spatial structures become increasingly complex to be analysed at a glance. Thus, representing and acquiring the required spatial information is more costly than expected and decision makers may get frustrated during the analysis. On the other hand, Web Engineering address similar problems (heterogeneous audience, different data sources and increasing amount and complexity of information) by using personalization rules. PRML (Personalization Rules Modeling Language) is a language that has been successfully applied to several Web systems in order to perform those personalization rules for every particular user and needs. Therefore, we have decided to use personalization rules and we have adapted the PRML to certain SDW aspects in order to introduce the right spatiality and deliver the correct information for every user needs. The great advantage of our approach is that each decision maker can easily include spatial data according to their own needs at conceptual level, while they can also conceptually get the right spatial schema instance avoiding exploring in a large and complex SDW.