On Ranking and Choice Models

S. Agarwal
2016-07-09
Abstract:In today's big data era, huge amounts of ranking and choice data are generated on a daily basis, and consequently, many powerful new computational tools for dealing with ranking and choice data have emerged in recent years. This paper highlights recent developments in two areas of ranking and choice modeling that cross traditional boundaries and are of multidisciplinary interest: ranking from pairwise comparisons, and automatic discovery of latent categories from choice survey data.
What problem does this paper attempt to address?