Lattice-Based Approaches for Enumerating Stable Extensions in Abstract Argumentation Frameworks

From International Center for Computational Logic

Toggle side column

Lattice-Based Approaches for Enumerating Stable Extensions in Abstract Argumentation Frameworks

Sergei ObiedkovSergei Obiedkov,  Barış SertkayaBarış Sertkaya
Lattice-Based Approaches for Enumerating Stable Extensions in Abstract Argumentation Frameworks


Sergei Obiedkov, Barış Sertkaya
Lattice-Based Approaches for Enumerating Stable Extensions in Abstract Argumentation Frameworks
In Madalina Croitoru, Domingo López-Rodríguez, Gerd Stumme, eds., The Third International Joint Conference on Conceptual Knowledge Structures (CONCEPTS 2026), LNCS, to appear. Springer
  • KurzfassungAbstract
    We investigate two approaches for computing stable extensions in abstract argumentation frameworks. The first represents stable extensions using Formal Concept Analysis (FCA), where they form an antichain in the concept lattice of a formal context derived from the framework. To improve the performance of this approach on sparse frameworks, we combine it with a dynamic-programming strategy based on decomposition into strongly connected components. The second approach transforms the attack relation into an implicational system such that the complements of stable extensions are among its models. FCA algorithms can then be used to explore the corresponding lattice and enumerate stable extensions. We experimentally evaluate the proposed techniques.
  • Projekt:Project: CPECSECAIScaDS.AI
  • Forschungsgruppe:Research Group: Wissensbasierte SystemeKnowledge-Based Systems
The final publication is available at Springer.
@inproceedings{OS2026,
  author    = {Sergei Obiedkov and Bar{\i}{\c{s}} Sertkaya},
  title     = {Lattice-Based Approaches for Enumerating Stable Extensions in
               Abstract Argumentation Frameworks},
  editor    = {Madalina Croitoru and Domingo L{\'{o}}pez-Rodr{\'{\i}}guez and
               Gerd Stumme},
  booktitle = {The Third International Joint Conference on Conceptual Knowledge
               Structures (CONCEPTS 2026)},
  series    = {LNCS},
  publisher = {Springer},
  year      = {2026}
}