Machine Translation for home and business users
291Authors
nan
Abstract
: Based on the long-standing experience with the project Logic-Programm- ing based Machine Translation (LMT), IBM and the publisher von Rheinbaben & Busch have developed Personal Translator®, a machine translation system which is geared toward the needs of home and business users. From the point of view of technology, the challenge has been to make most of the functionality available to general users which was originally developed in a research environment and with professional users in mind. It will be explained how the market of MT for home and business users was perceived and what features were deemed necessary for such a system. The technology of LMT which in its core parts originates from M. McCord (IBM Research) will be outlined. It will then be shown how general users can be enabled to tailor a system such as Personal Translator to their needs, in particular how they can modify the vocabulary. A discussion of translation quality and its parameters will follow. Finally, an outlook will be given of what can be expected in future MT systems.
Keywords
Full Content
Machine Translation for Home and Business Users Hubert Lehmann IBM Deutschland Informationssysteme GmbH Wissenschaftliches Zentrum Heidelberg hlehmann@vnet.ibm.com Abstract: Based on the long-standing experience with the project Logic-Programm- ing based Machine Translation (LMT), IBM and the publisher von Rheinbaben & Busch have developed Personal Translator®, a machine translation system which is geared toward the needs of home and business users. From the point of view of technology, the challenge has been to make most of the functionality available to general users which was originally developed in a research environment and with professional users in mind. It will be explained how the market of MT for home and business users was perceived and what features were deemed necessary for such a system. The technology of LMT which in its core parts originates from M. McCord (IBM Research) will be outlined. It will then be shown how general users can be enabled to tailor a system such as Personal Translator to their needs, in particular how they can modify the vocabulary. A discussion of translation quality and its parameters will follow. Finally, an outlook will be given of what can be expected in future MT systems. Introduction When research on machine translation started in the late forties, the linguistic theory which would have been needed as a foundation for such an ambitious undertaking simply did not exist. Since then progress in linguistics and in Natural Language Processing has been immense, even though we still only partially understand the problem. When Michael McCord of IBM Research started the project Logic-programming based Machine Translation (LMT) in 1985, he was able to build on the latest insights from linguistics, logic, and logic programming. As a result we now have a machine translation system which exhibits the most modern technology commercially available today. Research on LMT soon became an international effort with close and enthusiastic cooperation of IBM researchers in the USA, Denmark, Egypt, France, Germany, Israel, Italy, Japan, and Spain. Accordingly, a number of prototype MT systems were developed and tested with the translation of computer manuals and other types of texts. It could be shown that LMT was able to meet the need for machine translation, and a lot was learned about the requirements of different types of users and different types of translation tasks. The development of a commercially viable product based on the existing prototype software has been mainly the responsibility of IBM Germany's Heidelberg Science Center, and it was done in cooperation with the publisher von Rheinbaben & Busch, who has been closely involved in the product design and now is responsible for the marketing of Personal Translator. In the sections to follow the topics will be: the market for machine translation, requirements for a machine translation system, the technology of Personal Translator, its user interface, a discussion of translation quality. In closing, an outlook will be given of how the development of MT systems may continue, and what can be expected for the years to come. The market for machine translation …
Content truncated to 500 words
Labeled Text
: Based on the long-standing experience with the project Logic-Programm- ing based Machine Translation (LMT), IBM and the publisher von Rheinbaben & Busch have developed Personal Translator®, a machine translation system which is geared toward the needs of home and business users. From the point of view of technology, the challenge has been to make most of the functionality available to general users which was originally developed in a research environment and with professional users in mind. It will be explained how the market of MT for home and business users was perceived and what features were deemed necessary for such a system. The technology of LMT which in its core parts originates from M. McCord (IBM Research) will be outlined. It will then be shown how general users can be enabled to tailor a system such as Personal Translator to their needs, in particular how they can modify the vocabulary. A discussion of translation quality and its parameters will follow. Finally, an outlook will be given of what can be expected in future MT systems.
Publication Details
- Paper ID:
- 291
- Year:
- 1995
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan