PP-2012-18: Extracting tree fragments in linear average time

PP-2012-18: van Cranenburgh, Andreas (2012) Extracting tree fragments in linear average time. [Report]

Text (Full Text)

Download (172kB) | Preview
[img] Text (Abstract)

Download (523B)


This report details the implementation of a fragment extraction algorithm using an average case linear time tree kernel. Given a treebank, the algorithm extracts all fragments that occur at least twice, along with their frequency. Evaluation shows a 70-fold speedup over a quadratic fragment extraction implementation. Additionally, we add support for trees with discontinuous constituents.

Item Type: Report
Report Nr: PP-2012-18
Series Name: Prepublication (PP) Series
Year: 2012
Uncontrolled Keywords: treebanks; fragments; tree kernels
Depositing User: Andreas Cranenburgh
Date Deposited: 12 Oct 2016 14:37
Last Modified: 12 Oct 2016 14:37
URI: https://eprints.illc.uva.nl/id/eprint/460

Actions (login required)

View Item View Item