Abductive Logic Programming - Comparison With Other Frameworks

Comparison With Other Frameworks

ALP can be viewed as a refinement of Theorist, which explored the use of abduction in first-order logic for both explanation and default reasoning. David Poole later developed a logic programming variant of Theorist, in which abducible predicates have associated probabilities. He showed that probabilistic Horn abduction incorporates both pure Prolog and Bayesian networks as special cases.

There exist strong links between some ALP frameworks and other extensions of logic programming. In particular, ALP has close connections with Answer Set Programming. An abductive logic program can be translated into an equivalent answer set program under the stable model semantics. Consequently, systems for computing stable models such as SMODELS can be used to compute abduction in ALP.

ALP is also closely related to constraint logic programming (CLP). On the one hand, the integration of constraint solving and abductive logic programming enhances the practical utility of ALP through a more efficient computation of abduction. On the other hand, the integration of ALP and CLP can be seen as a high-level constraint programming environment that allows more modular and flexible representations of the problem domain.

There is also a strong link between ALP and Argumentation in logic programming. This relates both to the interpretation of Negation as Failure and Integrity Constraints.