PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Subtree Testing and Closed Tree Mining Through Natural Representations
José Balcázar, Albert Bifet and Antoni Lozano
In: ACKE'07: International Workshop on Advances in Conceptual Knowledge Engineering, 3-7 September 2007, Regensburg, Germany.

Abstract

Several classical schemes exist to represent trees as strings over a fixed alphabet; these are useful in many algorithmic and conceptual studies. Our previous work has proposed a representation of unranked trees as strings over a countable alphabet, and has shown how this representation is useful for canonizing unordered trees and for mining closed frequent trees, whether ordered or unordered. Here we propose a similar, simpler alternative and adapt some basic algorithmics to it; then we show empirical evidence of the usefulness of this representation for mining frequent closed unordered trees on real-life data.

PDF - PASCAL Members only - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:3224
Deposited By:Albert Bifet
Deposited On:21 January 2008