PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Dryade: a new approach for discovering closed frequent trees in heterogeneous tree databases
Alexandre Termier, Marie-Christine Rousset and Michele Sebag
In: ICDM 2004, IEEE Int. Conference on Data Mining, november 2004, Brighton, UK.

Abstract

This paper presents a new algorithm for discovering frequent tree patterns in a tree database. A relaxed tree inclusion definition is used in order to accommodate heterogeneous databases, resulting in an NP complete computational complexity. Good performances are obtained by i) delegating the construction and evaluation of candidate patterns to propositional frequent itemset algorithms., ii) restricting the search to closed frequent tree patterns.

Postscript - Requires a viewer, such as GhostView
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:611
Deposited By:Michele Sebag
Deposited On:29 December 2004