Multi-Way Block Models

This dissertation investigates the Multi-Way Block Models proposed for mining inner structure of interaction between two sets of subjects. The Multi-Way Block Models generalize the Mixed Membership Stochastic Block Models (Airoldi et al., 2008) in multi-ways, as extensions in model settings allowing for different distributions of interactions between two groups of subjects. Moreover, the Multi-Way Block Models generalize model implementations to variational Bayesian, collapsed Gibbs sampling, collapsed variational Bayesian, and expectation propagation approaches. Comparative simulation studies show that the four implementation algorithms achieve meaningful parameter estimates for the latent membership and block structure from correlation and network among the subjects.

Name thesis:degree_name

PhD

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Arts and Sciences: Mathematical Sciences

Grantor dc:publisher

University of Cincinnati

Year dc:date

2012

Author dc:creator

• Wang, Xiaopei

Contributors dc:contributor

• Deddens, James

• Statistics

Statement dc:rights

• unrestricted
• This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.

Language dc:language

English

Repository record dc:identifier

http://rave.ohiolink.edu/etdc/view?acc_num=ucin1342716695

OAI identifier oai:identifier

oai:etd.ohiolink.edu:ucin1342716695