Towards Multi-Agent Knowledge Allocation - LIRMM - Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier
Reports Year : 2012

Towards Multi-Agent Knowledge Allocation

Abstract

The classical setting of query answering either assumes the existence of just one knowledge requester, or the knowledge requests from different parties are treated independently from each other. This assumption does not always hold in practical applications where requesters often are in direct competition for knowledge. We provide a formal model for this type of scenario scenario by proposing the Multi-Agent Knowledge Allocation (MAKA) setting which combines the fields of query answering in information systems and multi-agent resource allocation. We define a bidding language based on exclusivity-annotated conjunctive queries and succinctly translate the allocation problem into a graph structure allowing to employ a wide range of constraint solving techniques for optimal allocation.
Fichier principal
Vignette du fichier
technical_report-MAKA.pdf (194.87 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

lirmm-00657422 , version 1 (06-01-2012)
lirmm-00657422 , version 2 (17-01-2012)

Identifiers

  • HAL Id : lirmm-00657422 , version 2
  • PRODINRA : 245734

Cite

Madalina Croitoru, Sebastian Rudolph. Towards Multi-Agent Knowledge Allocation. RR-12003, 2012, pp.18. ⟨lirmm-00657422v2⟩
192 View
264 Download

Share

More