Graphical interaction with heterogeneous databases

During the past few years our research efforts have been inspired by two different needs. On one hand, the number of non-expert users accessing databases is growing apace. On the other, information systems will no longer be characterized by a single centralized architecture, but rather by several heterogeneous component systems. In order to address such needs we have designed a new query system with both user-oriented and multidatabase features. The system's main components are an adaptive visual interface, providing the user with different and interchangeable interaction modalities, and a "translation layer", which creates and offers to the user the illusion of a single homogeneous schema out of several heterogeneous components. Both components are founded on a common ground, i.e. a formally defined and semantically rich data model, the Graph Model, and a minimal set of Graphical Primitives, in terms of which general query operations may be visually expressed. The Graph Model has a visual syntax, so that graphical operations can be applied on its components without unnecessary mappings, and an object-based semantics. The aim of this paper is twofold. We first present an overall view of the system architecture and then give a comprehensive description of the lower part of the system itself. In particular, we show how schemata expressed in different data models can be translated in terms of Graph Model, possibly by exploiting reverse engineering techniques. Moreover, we show how mappings can be established between well-known query languages and the Graphical Primitives. Finally, we describe in detail how queries expressed by using the Graphical Primitives can be translated in terms of relational expressions so to be processed by actual DBMSs.