You need JavaScript to view this

Semantics and artificial intelligence in machine translation

Journal Article:

Abstract

The author exemplifies three types of ambiguity that the introduction of semantics or of AI methods might be expected to solve: word sense, structural, and referential ambiguity. From this point of view she examines the works of Schank, Riesbeck, Minsky, Charniak, and Wilks, and she comes to the conclusion that the systems described will not be of much help for the development of operational MT-systems, except within a well-defined, constrained world. The latter aspect is illustrated by the author by means of a description of the Edinburgh Mecho-project. But, as the vast majority of texts destined for MT does not come from a constrained world, such systems will hardly be used as MT production systems. Still, MT-systems like Eurotra give the chance of making intelligent use of AI ideas. 16 references.
Authors:
Publication Date:
Jan 01, 1981
Product Type:
Journal Article
Reference Number:
EDB-85-152518
Resource Relation:
Journal Name: Sprache Datenverarb.; (Germany, Federal Republic of); Journal Volume: 1-2
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; MACHINE TRANSLATIONS; ARTIFICIAL INTELLIGENCE; 990200* - Mathematics & Computers
OSTI ID:
5363422
Research Organizations:
Univ. of Geneva, Switzerland
Country of Origin:
Germany
Language:
English
Other Identifying Numbers:
Journal ID: CODEN: SPDAD
Submitting Site:
HEDB
Size:
Pages: 5-8
Announcement Date:

Journal Article:

Citation Formats

King, M. Semantics and artificial intelligence in machine translation. Germany: N. p., 1981. Web.
King, M. Semantics and artificial intelligence in machine translation. Germany.
King, M. 1981. "Semantics and artificial intelligence in machine translation." Germany.
@misc{etde_5363422,
title = {Semantics and artificial intelligence in machine translation}
author = {King, M}
abstractNote = {The author exemplifies three types of ambiguity that the introduction of semantics or of AI methods might be expected to solve: word sense, structural, and referential ambiguity. From this point of view she examines the works of Schank, Riesbeck, Minsky, Charniak, and Wilks, and she comes to the conclusion that the systems described will not be of much help for the development of operational MT-systems, except within a well-defined, constrained world. The latter aspect is illustrated by the author by means of a description of the Edinburgh Mecho-project. But, as the vast majority of texts destined for MT does not come from a constrained world, such systems will hardly be used as MT production systems. Still, MT-systems like Eurotra give the chance of making intelligent use of AI ideas. 16 references.}
journal = {Sprache Datenverarb.; (Germany, Federal Republic of)}
volume = {1-2}
journal type = {AC}
place = {Germany}
year = {1981}
month = {Jan}
}