Incremental Computation in Dynamic Argumentation Frameworks
Gianvincenzo Alfano, Sergio Greco, Francesco Parisi · IEEE Intelligent Systems · 2021
Dealing with controversial information is a challenging and important task for intelligent systems. Formal argumentation enables reasoning on arguments for and against a claim to decide on an outcome. An argumentation framework often models a dynamic situation where arguments as well as the way they interact frequently change over the time. As a consequence, the sets of accepted arguments (i.e., extensions under a given semantics) often need to be computed again after performing an update. In this article, we address the problem of efficiently recomputing extensions of dynamic argumentation frameworks. We present an incremental algorithmic solution whose main idea is that of using an initial extension and the update to identify a (potentially small) portion of the argumentation framework, which is sufficient to compute an extension of the whole updated framework.