Showing papers by "Antonio Brogi published in 1990"

Composition Operators for Logic Theories

Abstract: Some basic meta-level operators for putting logic theories together are introduced, which relate to set-theoretic union, intersection and difference. Both a transformational and an interpretive characterization of the operators are provided and proved equivalent. The former definition says how to syntactically construct a new theory out of two given theories, the latter provides a meta-level interpretation of the same operators. A declarative — both model-theoretic and fixpoint — semantics of the operators is also provided, allowing one to assign the minimal model of the resulting theory as a function of the models of the argument theories. Some examples from default reasoning, knowledge assimilation, inheritance networks and hypothetical reasoning are presented to demonstrate the expressive power of the operators.

Objects in a Logic Programming Framework

Abstract: Some basic notions of object-oriented programming such as objects, messages and inheritance are provided of a clean definition according to the logic programming paradigm Objects are represented by logic theories, while inheritance mechanisms are expressed through meta-level axioms The sending of a message is interpreted as a request for the proof of a formula The main contribution of the paper is to provide a clean semantic characterization for these notions defining both a proof-theory and a compositional model-theory

Logic Programming within a Functional Framework

Abstract: The functional programming language LML (for Logical Meta-Language) is presented. Like most trendy representatives of its category, LML is a higher-order, pattern-matched, polymorphically-typed, non-strict functional language. Its distinctive feature is the presence of a data type of theories, whose objects represent logic programs. Theories are equipped with operators for querying them, to obtain sets of answers, and combining them together in various different ways, to build more complex ones. Thus, from one perspective, LML may be viewed as yet another approach to the integration of functional and logic programming, aiming at amalgamating within a single, coherent framework the expressive power of both paradigms. From another perspective, however, LML may be viewed as a programming language for the construction of knowledge-based systems, in which chunks of knowledge are represented as logic theories. According to this perspective, the functional layer acts as a meta-level framework for the underlying logic programming component: theories are ordinary data values, which can be manipulated by suitable operators. The operators map theories into theories by acting upon their representations according to given formal semantics. This is the most novel aspect of the language, and provides mechanisms for a) the description of dynamic evolution of theories, and b) a modular approach to knowledge-based systems development. The paper presents the basic ideas underlying the design of the language, together with anexample of its use.