Title :
Moving toward semantics for language processing: recent advances in resource construction and application
Author_Institution :
USC Inf. Sci. Inst., Marina del Rey, CA, USA
fDate :
Oct. 30 2005-Nov. 1 2005
Abstract :
Summary form only given. In this paper the author describes the recent work at various locations, focusing on the resource components required (including ISI´s symbol definition ontology Omega and the BBN-UPenn-Colorado-ISI large corpus OntoBank of (shallow) meaning representations) and the resources and methods one needs to build them (including existing ontologies, human annotation procedures, and a verification methodology, as embodied in a pilot project called Learning by Reading that involves various prominent researchers). This work is bringing together the NLP and KE/KR communities after a separation of some 40 years, and poses challenges for each side: dealing with large amounts of possibly incorrect knowledge is not something KE/KR has comfortable with, and dealing with requirements that knowledge be formalized and consistent is not something that large-scale NLP has been comfortable with. The research discussed here enables numerous experiments on the wide-scale use of (shallow) semantics, to the benefit of NLP applications of all kinds.
Keywords :
natural languages; ontologies (artificial intelligence); programming language semantics; NLP; knowledge engineering; knowledge representation; natural language processing; ontologies; semantics; Data mining; Handicapped aids; Humans; Intersymbol interference; Large-scale systems; Natural language processing; Ontologies; Statistical analysis;
Conference_Titel :
Natural Language Processing and Knowledge Engineering, 2005. IEEE NLP-KE '05. Proceedings of 2005 IEEE International Conference on
Conference_Location :
Wuhan, China
Print_ISBN :
0-7803-9361-9
DOI :
10.1109/NLPKE.2005.1598695