Computing utility from weighted description logic preference formulas

  • Authors:
  • Azzurra Ragone;Tommaso Di Noia;Francesco M. Donini;Eugenio Di Sciascio;Michael P. Wellman

  • Affiliations:
  • SisInfLab, Politecnico di Bari, Bari, Italy;SisInfLab, Politecnico di Bari, Bari, Italy;Università della Tuscia, Viterbo, Italy;SisInfLab, Politecnico di Bari, Bari, Italy;Artificial Intelligence Laboratory, University of Michigan, Ann Arbor

  • Venue:
  • DALT'09 Proceedings of the 7th international conference on Declarative Agent Languages and Technologies
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a framework to compute the utility of a proposal w.r.t. a preference set in a negotiation process. In particular, we refer to preferences expressed as weighted formulas in a decidable fragment of First Order Logic (FOL). Although here we tailor our approach for Description Logics endowed with disjunction, all the results keep their validity in any decidable fragment of FOL. DLs offer expressivity advantages over propositional representations, and allow us to relax the often unrealistic assumption of additive independence among attributes. We provide suitable definitions of the problem and present algorithms to compute utility in our setting. We also study complexity issues of our approach and demonstrate its usefulness with a running example in a multiattribute negotiation scenario.