Abstract Details
Activity Number:
|
658
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract #311761
|
View Presentation
|
Title:
|
Poisson Hierarchical Biclustering Model
|
Author(s):
|
Thao Duong*+
|
Companies:
|
Donald Bren School of Information and Computer Science
|
Keywords:
|
bayesian ;
hierarchical ;
clustering ;
biclustering ;
poisson ;
count data
|
Abstract:
|
Biclustering refers to identification of subsets of individuals (units) and subsets of measurements (features) that define interesting partitions of the data. In this paper,we present a hierarchical Bayesian methodology for biclustering count data via a log link. We begin with a model for a known number of biclusters. We handle the problem of identifying the discriminative features and the observations in each cluster by introducing binary latent vectors (for rows and columns separately) for each bicluster. The model is extended to incorporate the number of biclusters as unknown parameter. The resulting method selects the optimal number of clusters, discriminating features and observations in each group simultaneously. We apply these methodologies to simulated data and Simons Facial Dysmorphology Measurements. For this model we mainly use Markov Chain Monte Carlo techniques to simulate samples from the conditional posterior distributions of interest. The samples are then utilized in making inferences from the models. In addition to full Bayesian analyses, the use of maximum likelihood for obtaining inferences is also described and applied.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.