Learning utility functions from preference relations on graphs

  • Géraldine Bous
Conference paper
Part of the Operations Research Proceedings book series (ORP)


The increasing popularity of graphs as fundamental data structures, due to their inherent flexibility in modeling information and its structure, has led to the development of methods to efficiently store, search and query graphs. Graphs are nonetheless complex entities whose analysis is cognitively challenging. This calls for the development of decision support systems that build upon a measure of ‘usefulness’ of graphs. We address this problem by introducing and defining the concept of ‘graph utility’. As the direct specification of utility functions is itself a difficult problem, we explore the problem of learning utility functions for graphs on the basis of user preferences that are extracted from pairwise graph comparisons.


Utility Function Recommendation System Marginal Utility Utility Model Query Graph 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bous, G., Fortemps, Ph., Glineur, F., Pirlot,M.: ACUTA: A novel method for eliciting additive value functions on the basis of holistic preference statements. EJOR 206, 435–444 (2010)CrossRefGoogle Scholar
  2. 2.
    Fishburn, P.C.: Utility Theory for Decision Making. Wiley, New York (1970)Google Scholar
  3. 3.
    Greco, S.,Mousseau, V., Słowi´nski, R.: Ordinal regression revisited: multiple criteria ranking using a set of additive value functions. EJOR 191, 415–435 (2008)Google Scholar
  4. 4.
    Huard, P.: Resolution of mathematical programming with nonlinear constraints by the method of centers. In: Abadie, J. (ed.) Nonlinear Programming, pp. 209–219.Wiley, New York (1967)Google Scholar
  5. 5.
    Jacquet-Lagr`eze, E., Siskos, Y.: Assessing a set of additive utility functions to multicriteria decision-making: the UTA method. EJOR 10, 151–164 (1982)Google Scholar
  6. 6.
    Keeney, R.L., Raiffa, H.: Decisions with multiple objectives: Preferences and value tradeoffs. Wiley, New York (1976)Google Scholar
  7. 7.
    Nesterov, Y.E., Nemirovskii, A.S.: Interior-point Polynomial Algorithms in Convex Programming. SIAM, Philadelphia (1994)CrossRefGoogle Scholar
  8. 8.
    Siskos, Y., Grigoroudis, E., Matsatsinis, N.: UTA Methods. In: Figueira, J., Greco, S., Ehrgott, M. (eds.) Multiple Criteria Decision Analysis: State of the Art Surveys, pp. 297–334. Springer, New York (2005)Google Scholar
  9. 9.
    Sonnevend, G.: An analytical centre for polyhedrons and new classes of global algorithms for linear (smooth, convex) programming. In: Prekopa, A., Szelezsan, J., Strazicky, B. (eds.) LNCIS, pp. 866–876. Springer, Heidelberg (1985)Google Scholar
  10. 10.
    Ye, Y.: Interior Point Algorithms: Theory and Analysis. Wiley, New York (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.SAP Research Sophia Antipolis, Business Intelligence PracticeMouginsFrance

Personalised recommendations