Toward an MT System without Pre-Editing --- Effects of New Methods in ALT-J/E ---
Satoru Ikehara, Satoshi Shirai, Akio Yokoo, Hiromi Nakaiwa
Abstract
Recently, several types of Japanese-to-English machine translation systems have been developed, but all of them require an initial process of rewriting the original text into easily translatable Japanese. Therefore these systems are unsuitable for translating information that needs to be speedily disseminated. To overcome this limitation, a Multi-Level Translation Method based on the Constructive Process Theory has been proposed. This paper describes the benefits of using this method in the Japanese-to-English machine translation system ALT-J/E. In comparison with conventional compositional methods, the Multi-Level Translation Method emphasizes the importance of the meaning contained in expression structures as a whole. It is shown to be capable of translating typical written Japanese based on the meaning of the text in its context, with comparative ease. We are now hopeful of carrying out useful machine translation with no manual pre-editing.
Create a lesson
Related papers
A Memory-Based Approach to Learning Shallow Natural Language Patterns
Shlomo Argamon, Ido Dagan, Yuval Krymolowski
A Comparison of WordNet and Roget's Taxonomy for Measuring Semantic Similarity
Michael Mc Hale
Some Ontological Principles for Designing Upper Level Lexical Resources
Nicola Guarino
Towards an implementable dependency grammar
Timo Jarvinen, Pasi Tapanainen
A Variant of Earley Parsing
Mark-Jan Nederhof, Giorgio Satta
Segregatory Coordination and Ellipsis in Text Generation
James Shaw