TreeFinder: a First Step towards XML Data Mining Alexandre Termier, Marie-Christine Rousset, and Michèle Sebag   View Paper (PDF)   Return to Main-Track Regular Papers Abstract In this paper, we consider the problem of searching fre-quent trees from a collection of tree-structured data model-ing XML data. The TreeF inder algorithm aims at finding trees, such that their exact or perturbed copies are frequent in a collection of labelled trees. To cope with complexity issues, TreeF inder is correct but not complete: it finds a subset of the actually frequent trees. The default of completeness is experimentally inves-tigated on artificial medium size datasets; it is shown that TreeFinderreaches completeness or falls short to it for a range of experimental settings. DiSC'03 © 2003 Association for Computing Machinery