Constant Factor Approximation of Edit Distance of Bounded Height Unordered Trees

  • Authors:
  • Daiji Fukagawa;Tatsuya Akutsu;Atsuhiro Takasu

  • Affiliations:
  • National Institute of Informatics, Tokyo, Japan 101---8430;Bioinformatics Center, Institute for Chemical Research, Kyoto University, Kyoto, Japan 611---0011;National Institute of Informatics, Tokyo, Japan 101---8430

  • Venue:
  • SPIRE '09 Proceedings of the 16th International Symposium on String Processing and Information Retrieval
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The edit distance problem on two unordered trees is known to be MAX SNP-hard. In this paper, we present an approximation algorithm whose approximation ratio is 2h + 2, where we consider unit cost edit operations and h is the maximum height of the two input trees. The algorithm is based on an embedding of unit cost tree edit distance into L 1 distance. We also present an efficient implementation of the algorithm using randomized dimension reduction.