Bayesian nonparametric models for ranked data

Journal article

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.

Authors: Caron, F; Teh, Y

• Journal: Advances in Neural Information Processing Systems
• Volume: 2
• Pages: 1520-1528
• Publication date: 2012-11-19
• ISSN: 1049-5258
• Language: English
• Keywords: stat.ME; stat.ML; cs.LG