Architecture of a relational decision support system

摘要

The architecture of the POLD relational system for distributed data analysis is presented. POLD was designed as the kernel of a decision support information system. The peculiarity of POLD consists in allowing the user to view and analyze the data information as couple 〈data, semantics〉, the semantics being a user's suitable classification of the data. This approach furnishes the user with a powerful tool for the semantic integration of data coming from different and/or distributed sources. Both semantics and user views are dynamically definable, that is, they may change with user, current time, and application. The POLD system supports an extended version of the relational operations in order to fit the proposed approach. Moreover, this system provides features for statistical analyses, reports and graphical representations.