A novel logic program like language, weight constraint rules, is developed for answer set programming purposes. It generalizes normal logic programs by allowing weight constraints in place of literals to represent, e.g., cardinality and resource constraints and by providing optimization capabilities. A declarative semantics is developed which extends the stable model semantics of normal programs. The computational complexity of the language is shown to be similar to that of normal programs under the stable model semantics. A simple embedding of general weight constraint rules to a small subclass of the language called basic constraint rules is devised. An implementation of the language, the smodels system, is developed based on this embedding. It uses a two level architecture consisting of a front-end and a kernel language implementation. The front-end allows restricted use of variables and functions and compiles general weight constraint rules to basic constraint rules. A major part of the work is the development of an ecient search procedure for computing stable models for this kernel language. The procedure is compared with and empirically tested against satis ability checkers and an implementation of the stable model semantics. It offers a competitive implementation of the stable model semantics for normal programs and attractive performance for problems where the new types of rules provide a compact representation.

This paper surveys various complexity results on different forms of logic programming. The main focus is on decidable forms of logic programming, in particular, propositional logic programming and datalog, but we also mention general logic programming with function symbols. Next to classical results on plain logic programming (pure Horn clause programs), more recent results on various important extensions of logic programming are surveyed. These include logic programming with different forms of negation, disjunctive logic programming, logic programming with equality, and constraint logic programming. The complexity of the unification problem is also addressed.

In this paper, we address the issue of how Gelfond and Lifschitz's answer set semantics for extended logic programs can be suitably modified to handle prioritized programs. In such programs an ordering on the program rules is used to express preferences. We show how this ordering can be used to define preferred answer sets and thus to increase the set of consequences of a program. We define a strong and a weak notion of preferred answer sets. The first takes preferences more seriously, while the second guarantees the existence of a preferred answer set for programs possessing at least one answer set. Adding priorities

Disjunctive Deductive Databases (DDDBs) --- function-free disjunctive logic programs with negation in rule bodies allowed --- have been recently recognized as a powerful tool for knowledge representation and commonsense reasoning. Much research has been spent on issues like semantics and complexity of DDDBs, but the important area of implementing DDDBs has been less addressed so far. However, a thorough investigation thereof is a basic requirement for building systems which render previous foundational work on DDDBs useful for practice. This paper presents the architecture of a DDDB system currently developed at TU Vienna in the FWF project P11580-MAT "A Query System for Disjunctive Deductive Databases".

The need for representing indefinite information led to disjunctive deductive databases, which also fertilized work on disjunctive logic programming. Based on this paradigm, the DLV system has been designed and implemented as a tool for declarative knowledge representation. In this paper, we focus on the usage of DLV for solving problems in a declarative manner and report on experiments that we have run on a suite of benchmark problems. We discuss how problems can be solved in a natural way using a "Guess&Check"-paradigm where solutions are guessed and verified by parts of the program. Furthermore, we describe various front-ends that can be used for solving problems in specific applications. The experiments show that due to the ongoing implementation efforts, which include fine-tuning of the underlying algorithms, successive and significant performance improvements have been achieved during the last two years. The results indicate that DLV is capable of solving some complex problems quite efficiently.

The Smodels system is one of the state-of-the-art implementations of stable model computation for normal logic programs. In order to enable more realistic applications, the basic modeling language of normal programs has been extended with new constructs including cardinality and weight constraints and corresponding implementation techniques have been developed. This paper summarizes the extensions that have been included in the system, demonstrates their use, provides basic application methodology, illustrates the current level of performance of the system, and compares it to state-of-the-art satis ability checkers.

We present an extension of language A-Prolog by consistency-restoring rules with preferences, give the semantics of the new language, CR-Prolog, and show how the language can be used to formalize various types of commonsense knowledge and reasoning.

A generalization of logic program rules is proposed where rules are built from weight constraints with type information for each predicate instead of simple literals. These kinds of constraints are useful for concisely representing different kinds of choices as well as cardinality, cost and resource constraints in combinatorial problems such as product configuration. A declarative semantics for the rules is presented which generalizes the stable model semantics of normal logic programs. It is shown that for ground rules the complexity of the relevant decision problems stays in NP. The fust implementation of the language handles a decidable subset where function symbols are not allowed. It is based on a new procedure for computing stable models for ground rules extending normal programs with choice and weight constructs and a compilation technique where a weight rule with variables is transformed to a set of such simpler ground rules.

This paper presents the Diagnosis Frontend of dlv, which is a knowledge representation system under development at the Technische Universität Wien. The kernel language of the system is an extension of disjunctive logic programming (DLP) by integrity constraints; it offers frontends to several advanced knowledge representation formalisms. The formal model of diagnosis employed in the frontend includes both abductive diagnosis (over DLP theories) and consistency-based diagnosis. For each of the two diagnosis modalities, generic diagnoses, single error diagnoses, and subset minimal diagnoses are considered. We illustrate the use of the frontend by showing the dlv encodings of several diagnosis problems. Thereafter, we discuss implementation issues. Diagnostic reasoning is implemented on the dlv engine through suitable translations of diagnostic problems into disjunctive logic programs, such that their stable models correspond to diagnoses. For the six kinds of diagnostic reasoning problems emerging from above, such reductions are provided

In this paper we suggest an architecture for a software agent which operates a physical device and is capable of making observations and of testing and repairing the device components. We present novel de nitions of the notions of symptom, candidate diagnosis, and diagnosis which are based on the theory of action language AL. The new de nitions allow one to give a simple account of the agent's behavior in which many of the agent's tasks are reduced to computing stable models of logic programs.