Search

This paper deals with the management of data with different integrity grades, represented by marked data. In databases with explicit markings of damaged data, which integrity constraints apply depends on the markings of the referenced data. Correct data conform to the full set of integrity constraints while for data that is either damaged, but inessential or approximate, some integrity constraints can be relaxed. The main goal of this paper is to provide a foundation for marked databases extending the relational algebra. Results provided in this paper are preliminary, but they provide a pragmatic and reasonable approach without sacrificing the theoretical foundation.

This paper describes a new kind of logic-based approach to domain knowledge modelling in information systems engineering. The approach is centred around eDDL DP , a formal but user-friendly language used for an integrated description of structural and behavioural concepts at the domain level. These concepts, called information meta-concepts, represent the external outlook of the essential components of real-world application domains under a minimalist viewpoint. eDDL DP is an analyst-oriented, epistemological concept language which allows automatic reasoning about a domain knowledge base composed of information meta-concepts. The eDDL DP language is founded on an extensible framework explicitly conceived for the development of a new generation of computer-based analysis tools endowed with automatic deductive capabilities. The syntax of the language as well as some of the automatic reasoning capabilities bound to eDDL DP descriptions are shown by means of a real-world example. The main features of the eDDL DP -centred approach are compared with the corresponding features of two popular structural and behavioural modelling approaches.