Bayesian Multidimensional Scaling and Choice of Dimension

Abstract

Multidimensional scaling is widely used to handle data which consist of dissimilarity measures between pairs of objects or people. We deal with two major problems in metric multidimensional scaling--configuration of objects and determination of the dimension of object configuration--within a Bayesian framework. A Markov chain Monte Carlo algorithm is proposed for object configuration, along with a simple Bayesian criterion for choosing their effective dimension, called MDSIC. Simulation results are presented, as well as examples on real data. Our method provides better results than classical multidimensional scaling for object configuration, and MDSIC seems to work well for dimension choice in the examples considered.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2000
Accession Number
ADA458817

Entities

People

  • Adrian Raftery
  • Man-suk Oh

Organizations

  • University of Washington

Tags

Communities of Interest

  • C4I

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Bayesian Networks
  • Computational Science
  • Data Science
  • Information Operations
  • Information Science
  • Instructions
  • Iterations
  • Markov Chains
  • Monte Carlo Method
  • Simulations
  • Statistics

Readers

  • Computer Vision.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms