FINDING APPROXIMATE LANGUAGE PATTERNS

Samuel W. K. Chan

Abstract

A two-phase annotation method for semantic labeling based on the edit distance is proposed. This dynamic programming approach stresses on a non-exact string matching technique that takes full advantage of the underlying grammatical structure of 65,000 parse trees in a Treebank. It is based on the assumption that human language understanding is relevant to concrete past language experiences rather than any abstract linguistic rules. This shallow technique is inspired by the research in the area of bio-molecular sequences analysis which advocates high sequence similarity usually implies significant function or structural similarity. Experimental results for recognizing various labels in 10,000 sentences are used to justify its significances.

Download


Paper Citation


in Harvard Style

W. K. Chan S. (2008). FINDING APPROXIMATE LANGUAGE PATTERNS . In Proceedings of the First International Conference on Bio-inspired Systems and Signal Processing - Volume 2: BIOSIGNALS, (BIOSTEC 2008) ISBN 978-989-8111-18-0, pages 295-301. DOI: 10.5220/0001063802950301


in Bibtex Style

@conference{biosignals08,
author={Samuel W. K. Chan},
title={FINDING APPROXIMATE LANGUAGE PATTERNS},
booktitle={Proceedings of the First International Conference on Bio-inspired Systems and Signal Processing - Volume 2: BIOSIGNALS, (BIOSTEC 2008)},
year={2008},
pages={295-301},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001063802950301},
isbn={978-989-8111-18-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Bio-inspired Systems and Signal Processing - Volume 2: BIOSIGNALS, (BIOSTEC 2008)
TI - FINDING APPROXIMATE LANGUAGE PATTERNS
SN - 978-989-8111-18-0
AU - W. K. Chan S.
PY - 2008
SP - 295
EP - 301
DO - 10.5220/0001063802950301