8. Discovering Nonbinary Hierarchical Structures with Bayesian Rose Trees

  1. Kerrie L. Mengersen3,
  2. Christian P. Robert4 and
  3. D. Michael Titterington5
  1. Charles Blundell1,
  2. Yee Whye Teh1 and
  3. Katherine A. Heller2

Published Online: 24 APR 2011

DOI: 10.1002/9781119995678.ch8

Mixtures: Estimation and Applications

Mixtures: Estimation and Applications

How to Cite

Blundell, C., Teh, Y. W. and Heller, K. A. (2011) Discovering Nonbinary Hierarchical Structures with Bayesian Rose Trees, in Mixtures: Estimation and Applications (eds K. L. Mengersen, C. P. Robert and D. M. Titterington), John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9781119995678.ch8

Editor Information

  1. 3

    School of Mathematical Sciences, Queensland University of Technology, Australia

  2. 4

    Université Paris-Dauphine, CEREMADE, Paris, France

  3. 5

    University of Glasgow, Glasgow, UK

Author Information

  1. 1

    Gatsby Computational Neuroscience Unit, University College London, UK

  2. 2

    Department of Engineering, University of Cambridge, UK

Publication History

  1. Published Online: 24 APR 2011
  2. Published Print: 15 APR 2011

ISBN Information

Print ISBN: 9781119993896

Online ISBN: 9781119995678

SEARCH

Keywords:

  • discovering nonbinary hierarchical structures - with Bayesian rose trees;
  • rich hierarchical structures - common across disciplines;
  • Bayesian hierarchical clustering and Bayesian rose trees - on same synthetic dataset;
  • dataset of three clusters of data items - and no further internal substructure;
  • Bayesian rose tree mixture models;
  • prior work, diverse range of methods - for hierarchical structure discovery;
  • important area for hierarchical structure discovery - phylogenetic tree relating multiple species;
  • rose trees, partitions and mixtures;
  • avoiding needless cascades;
  • greedy construction of Bayesian rose tree mixtures - and Bayesian rose tree approach based on model selection

Summary

This chapter contains sections titled:

  • Introduction

  • Prior work

  • Rose trees, partitions and mixtures

  • Avoiding needless cascades

  • Greedy construction of Bayesian rose tree mixtures

  • Bayesian hierarchical clustering, Dirichlet process models and product partition models

  • Results

  • Discussion

  • References