Varro: An Algorithm and Toolkit for Regular Structure Discovery in Treebanks

The Varro toolkit is a system for identifying and counting a major class of regularity in treebanks and annotated natural language data in the form of treestructures: frequently recurring unordered subtrees. This software has been designed for use in linguistics to be maximally applicable to actually existing treebanks and other stores of tree-structurable natural language data. It minimizes memory use so that moderately large treebanks are tractable on commonly available computer hardware. This article introduces condensed canonically ordered trees as a data structure for efficiently discovering frequently recurring unordered subtrees. 1 Credits This research is supported by the AMASS++ Project1 directly funded by the Institute for the Promotion of Innovation by Science and Technology in Flanders (IWT) (SBO IWT 060051).

• Type: Journal
• Year: 2010
• Where: COLING
• Authors: Scott Martens
• Keywords: COLING 2010, Computational Linguistics, Natural Language Data, Tree-structurable Natural Language, Unordered Subtrees