8670 articles  [version française]

hal-00730792, version 1

Bayesian nonparametric models for ranked data

Francois Caron (, http://www.math.u-bordeaux1.fr/~fcaron/) a12, Yee Whye Teh (http://www.stats.ox.ac.uk/~teh/) b3

NIPS - Neural Information Processing Systems (2012)

Abstract: We develop a Bayesian nonparametric extension of the popular Plackett-Luce choice model that can handle an infinite number of choice items. Our framework is based on the theory of random atomic measures, with the prior specified by a gamma process. We derive a posterior characterization and a simple and effective Gibbs sampler for posterior simulation. We develop a time-varying extension of our model, and apply it to the New York Times lists of weekly bestselling books.

  • a –  INRIA
  • b –  Oxford University
  • 1:  ALEA (INRIA Bordeaux - Sud-Ouest)
  • INRIA – Université de Bordeaux – CNRS : UMR5251
  • 2:  Institut de Mathématiques de Bordeaux (IMB)
  • CNRS : UMR5251 – Université Sciences et Technologies - Bordeaux I – Université Victor Segalen - Bordeaux II
  • 3:  Department of Statistics [Oxford]
  • University of Oxford
  • Domain : Statistics/Machine Learning
    Statistics/Methodology
  • Keywords : choice models – generalized Bradley-Terry model – Plackett-Luce model – gamma process – Markov Chain Monte Carlo
  • Internal note : RR-8140
 
  • hal-00730792, version 1
  • oai:hal.inria.fr:hal-00730792
  • From: 
  • Submitted on: Sunday, 18 November 2012 20:15:29
  • Updated on: Monday, 19 November 2012 08:40:54