Bayesian Plackett--Luce latent block models for ranked data
arXiv:2607.26949
The paper proposes a Bayesian latent block model that jointly clusters assessors and items for ranked data using a Plackett–Luce observation model, with inference via Gibbs sampling and split‑merge MCMC, and demonstrates its use on cancer gene‑expression ranking data.
Abstract
We introduce a Bayesian latent block model that jointly partitions assessors and items under a Plackett--Luce observation model. Assessors are assigned to clusters and items to blocks; items in a block share a common strength parameter within each assessor cluster, yielding a parsimonious co-clustering representation. Independent Gnedin priors infer and . Data augmentation gives conjugate Gibbs updates and a tractable MCMC sampler with split-merge moves. Simulations characterize recovery and posterior uncertainty as signal, ranking depth, and group balance vary. Applied to the cancer gene atlas (TCGA) pan-cancer top-500 gene-expression rankings, the model reveals tissue-driven sample structure while compressing gene-level heterogeneity into interpretable blocks. Rank-based GSEA of posterior gene scores supports biological interpretation.
53 pages, 12 figures