H. Poon, and P. Domingos. Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1, page 1--10. Stroudsburg, PA, USA, Association for Computational Linguistics, (2009)
Abstract
We present the first unsupervised approach to the problem of learning a semantic parser, using Markov logic. Our USP system transforms dependency trees into quasi-logical forms, recursively induces lambda forms from these, and clusters them to abstract away syntactic variations of the same meaning. The MAP semantic parse of a sentence is obtained by recursively assigning its parts to lambda-form clusters and composing them. We evaluate our approach by using it to extract a knowledge base from biomedical abstracts and answer questions. USP substantially outperforms TextRunner, DIRT and an informed baseline on both precision and recall on this task.
%0 Conference Paper
%1 poon2009
%A Poon, Hoifung
%A Domingos, Pedro
%B Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1
%C Stroudsburg, PA, USA
%D 2009
%I Association for Computational Linguistics
%K domingos fol logic mln parsing poon semantic unsupervised
%P 1--10
%T Unsupervised Semantic Parsing
%U http://dl.acm.org/citation.cfm?id=1699510.1699512
%X We present the first unsupervised approach to the problem of learning a semantic parser, using Markov logic. Our USP system transforms dependency trees into quasi-logical forms, recursively induces lambda forms from these, and clusters them to abstract away syntactic variations of the same meaning. The MAP semantic parse of a sentence is obtained by recursively assigning its parts to lambda-form clusters and composing them. We evaluate our approach by using it to extract a knowledge base from biomedical abstracts and answer questions. USP substantially outperforms TextRunner, DIRT and an informed baseline on both precision and recall on this task.
%@ 978-1-932432-59-6
@inproceedings{poon2009,
abstract = {We present the first unsupervised approach to the problem of learning a semantic parser, using Markov logic. Our USP system transforms dependency trees into quasi-logical forms, recursively induces lambda forms from these, and clusters them to abstract away syntactic variations of the same meaning. The MAP semantic parse of a sentence is obtained by recursively assigning its parts to lambda-form clusters and composing them. We evaluate our approach by using it to extract a knowledge base from biomedical abstracts and answer questions. USP substantially outperforms TextRunner, DIRT and an informed baseline on both precision and recall on this task.},
acmid = {1699512},
added-at = {2014-01-05T16:21:32.000+0100},
address = {Stroudsburg, PA, USA},
author = {Poon, Hoifung and Domingos, Pedro},
biburl = {https://www.bibsonomy.org/bibtex/20ca061a3a5cc2782ebe6e0fb7a668f7d/jil},
booktitle = {Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1},
interhash = {49c8aa7e62d2904fa2166efa72c93472},
intrahash = {0ca061a3a5cc2782ebe6e0fb7a668f7d},
isbn = {978-1-932432-59-6},
keywords = {domingos fol logic mln parsing poon semantic unsupervised},
location = {Singapore},
numpages = {10},
pages = {1--10},
publisher = {Association for Computational Linguistics},
series = {EMNLP '09},
timestamp = {2015-12-20T21:37:11.000+0100},
title = {Unsupervised Semantic Parsing},
url = {http://dl.acm.org/citation.cfm?id=1699510.1699512},
year = 2009
}