The stable marriage problem has many practical applications in twosided markets like those that assign doctors to hospitals, students to schools, or buyers to vendors. Most algorithms to find stable marriages assume that the participants explicitly expresses a preference ordering. This can be problematic when the number of options is large or has a combinatorial structure.We consider therefore using CP-nets, a compact preference formalism in stable marriage problems. We study the impact of this formalism on the computational complexity of stable marriage procedures, as well as on the properties of the solutions computed by these procedures. We show that it is possible to model preferences compactly without significantly increasing the complexity of stable marriage procedures and whilst maintaining the desirable properties of the matching returned.

Compact Preference Representation in Stable Marriage Problems.

ROSSI, FRANCESCA;VENABLE, KRISTEN BRENT;
2009

Abstract

The stable marriage problem has many practical applications in twosided markets like those that assign doctors to hospitals, students to schools, or buyers to vendors. Most algorithms to find stable marriages assume that the participants explicitly expresses a preference ordering. This can be problematic when the number of options is large or has a combinatorial structure.We consider therefore using CP-nets, a compact preference formalism in stable marriage problems. We study the impact of this formalism on the computational complexity of stable marriage procedures, as well as on the properties of the solutions computed by these procedures. We show that it is possible to model preferences compactly without significantly increasing the complexity of stable marriage procedures and whilst maintaining the desirable properties of the matching returned.
2009
Algorithmic Decision Theory, First International Conference, ADT 2009, Proceedings
First International Conference, ADT 2009
9783642044274
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11577/2374204
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 9
  • ???jsp.display-item.citation.isi??? 4
  • OpenAlex ND
social impact