In this paper, we study efficient computation of tree similarity for ordered trees based on compressed subtree enumeration. The compressed subtree enumeration is a new paradigm of enumeration algorithms that enumerates all subtrees of an input tree T in the form of their compressed bit signatures. For the task of enumerating all compressed bit signatures of k-subtrees in an ordered tree T, we first present an enumeration algorithm in O(k)-delay, and then, present another enumeration algorithm in constant-delay using O(n) time preprocessing that directly outputs bit signatures. These algorithms are designed based on bit-parallel speed-up technique for signature maintenance. By experiments on real and artificial datasets, both algorithms showed approximately 22% to 36% speed-up over the algorithms without bit-parallel signature maintenance. © 2013 Springer-Verlag.
CITATION STYLE
Wasa, K., Hirata, K., Uno, T., & Arimura, H. (2013). Faster algorithms for tree similarity based on compressed enumeration of bounded-sized ordered subtrees. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8199 LNCS, pp. 73–84). https://doi.org/10.1007/978-3-642-41062-8_8
Mendeley helps you to discover research relevant for your work.