Guided forest edit distance: Better structure comparisons by using domain-knowledge

Abstract

We introduce the guided forest edit distance problem, which measures the similarity of two forests under the guidance of a third forest. We give an efficient algorithm for the problem. Our problem is a natural generalization of many important structure comparison problems such as the forest edit distance problem, constrained sequence alignment problem and the longest constrained common subsequence problem. Our algorithm matches the performance of the best known algorithms for these problems. © Springer-Verlag Berlin Heidelberg 2007.