Random Partition Models for Microclustering Tasks
From MaRDI portal
Abstract: Traditional Bayesian random partition models assume that the size of each cluster grows linearly with the number of data points. While this is appealing for some applications, this assumption is not appropriate for other tasks such as entity resolution, modeling of sparse networks, and DNA sequencing tasks. Such applications require models that yield clusters whose sizes grow sublinearly with the total number of data points -- the microclustering property. Motivated by these issues, we propose a general class of random partition models that satisfy the microclustering property with well-characterized theoretical properties. Our proposed models overcome major limitations in the existing literature on microclustering models, namely a lack of interpretability, identifiability, and full characterization of model asymptotic properties. Crucially, we drop the classical assumption of having an exchangeable sequence of data points, and instead assume an exchangeable sequence of clusters. In addition, our framework provides flexibility in terms of the prior distribution of cluster sizes, computational tractability, and applicability to a large number of microclustering tasks. We establish theoretical properties of the resulting class of priors, where we characterize the asymptotic behavior of the number of clusters and of the proportion of clusters of a given size. Our framework allows a simple and efficient Markov chain Monte Carlo algorithm to perform statistical inference. We illustrate our proposed methodology on the microclustering task of entity resolution, where we provide a simulation study and real experiments on survey panel data.
Recommendations
- Nonexchangeable random partition models for microclustering
- Clustering via nonsymmetric partition distributions
- Random partition models with regression on covariates
- A predictive view of Bayesian clustering
- Random partition models and exchangeability for Bayesian identification of population struc\-ture
Cites work
- scientific article; zbMATH DE number 2042826 (Why is no real title available?)
- scientific article; zbMATH DE number 774881 (Why is no real title available?)
- scientific article; zbMATH DE number 3297798 (Why is no real title available?)
- A Bayesian procedure for file linking to analyze end-of-life medical costs
- A hierarchical Bayesian approach to record linkage and population size problems
- A problem of the Allocation of Particles in Cells and Cycles of Random Permutations
- Characterizations of exchangeable partitions and random discrete distributions by deletion properties
- Combinatorial stochastic processes. Ecole d'Eté de Probabilités de Saint-Flour XXXII -- 2002.
- Consistency under sampling of exponential random graph models
- Detecting duplicates in a homicide registry using a Bayesian partitioning approach
- Edge exchangeable models for interaction networks
- Entity resolution with empirically motivated priors
- Informed proposals for local MCMC in discrete spaces
- Nonexchangeable random partition models for microclustering
- Random partition models and complementary clustering of Anglo-Saxon place-names
- Slice sampling. (With discussions and rejoinder)
- The Representation of Partition Structures
- Theoretical limits of microclustering for record linkage
Cited in
(14)- Clustering via nonsymmetric partition distributions
- Bayesian estimation of cluster covariance matrices of unknown form
- Random partition models and complementary clustering of Anglo-Saxon place-names
- Cohesion and Repulsion in Bayesian Distance Clustering
- Nonexchangeable random partition models for microclustering
- Entropy regularization in probabilistic clustering
- A Bayesian record linkage model incorporating relational data
- Graph-Aligned Random Partition Model (GARP)
- Fast Generation of Exchangeable Sequence of Clusters Data
- Interfaces for random cluster models
- Multifile Partitioning for Record Linkage and Duplicate Detection
- A Practical Approach to Proper Inference with Linked Data
- A Bayesian record linkage approach to applications in tree demography using overlapping LiDAR scans
- Bayesian learning of graph substructures
This page was built for publication: Random Partition Models for Microclustering Tasks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5881129)
