Relational Information Exchange and Aggregation in Multi-Context Systems

Information Flow within Relational Multi-context Systems

“…Examples are managed MCS [4], whose bridge rules allow arbitrary operations (e.g. deletion or revision operators) on context knowledge bases to be freely defined; relational MCSs [16], which introduce variables and aggregate expressions in bridge rules, extending the semantics of MCSs accordingly; or dynamic MCSs [11], designed to cope with situations where knowledge sources and their contents may change over time and are not known a priori. We will work within relational MCSs (defined formally in the next section), and discuss a possible generalization of dynamic MCSs at the end of Section 4.…”

“…MCSs have been a topic of active research in recent years, with effort being put in addressing specific problems of this framework, and several variants of MCSs have been proposed to deal with particular situations. Of particular interest are relational multi-context systems [16], where each context is allowed to have a first-order sublanguage freely generated from a given set of predicate symbols and constants. Relational MCSs generalize MCSs, rather than restricting them, since one can take the first-order sublanguage to be empty.…”

The IMPL Policy Language for Managing Inconsistency in Multi-Context Systems

“…In our running example we use bridge rules with variables, here we disregard the issue of instantiating these rules [16]. We denote by r[X|c] the ground version of bridge rule r where variable X has been substituted by constant c. Explaining Inconsistency in MCSs.…”

Semantic Reasoning with SPARQL in Heterogeneous Multi-context Systems

“…An existing proposal to support reasoning with variables in bridge rules of ordinary MCSs extends logics in a bottom-up way to 'relational logics' [8] by adding an extra set of predicate symbols and variables that can be used in bridge rules. Different from that, our approach is top-down as we introduce an intermediate variable mapping layer, which does not impose any restrictions on a context's logic or syntax.…”

“…These issues are not addressed in a satisfactory way by existing proposals to add variables to MCSs: in [7], variables are seen as schematic variables only, i.e., a bridge rule is only a short notation for the set of instantiated (thus variable-free) bridge rules. In [8], bridge rules with variables require that the contexts of the MCS adhere to a given syntax for predicates; and similar in [13] where only first order languages are considered. The first solution is infeasible for large numbers of instantiations, while the second and third cannot capture formalisms that do not use predicate syntax, e.g., RDF stores.…”