A practical algorithm for finding optimal triangulations

  • Authors:
  • Kirill Shoikhet;Dan Geiger

  • Affiliations:
  • Computer Science Department, Technion, Haifa, Israel;Computer Science Department, Technion, Haifa, Israel

  • Venue:
  • AAAI'97/IAAI'97 Proceedings of the fourteenth national conference on artificial intelligence and ninth conference on Innovative applications of artificial intelligence
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

An algorithm called QUICKTREE is developed for finding a triangulation T of a given undirected graph G such that the size of T's maximal clique is minimum and such that no other triangulation of G is a subgraph of T. We have tested QUICKTREE on graphs of up to 100 nodes for which the maximum clique in an optimal triangulation is of size 11. This is the first algorithm that can optimally triangulate graphs of such size in a reasonable time frame. This algorithm is useful for constraint satisfaction problems and for Bayesian inference through the clique tree inference algorithm.