Machine translation - evolution not revolution

MACHINE TRANSLATION - EVOLUTION NOT REVOLUTION - Jennifer A. Brundage SAP AG P.O. Box 14 61, D-69185 Walldorf Germany jennifer.brundage@sap.com Abstract The continuous trend towards globalization means that even the most modern of industries must constantly re-evaluate its strategies and adapt to new technologies. This not only involves living up to the demands set by the product life cycles but also to find solutions satisfying additional internal needs. As a long-time supporter of MT and TM technology, SAP has shown that it can make productive use of competitive, commercial NLP products. As a first step, an integrated solution using TM together with MT was targeted. Having implemented different solutions for two types of documentation, the focus is now on not merely to integrate other technologies (e.g. terminology mining or controlled language) but to provide a uniform solution for processing any type of text. This involves not only supporting the needs of technical writers and translators but of all employees in their multilingual working environment. Keywords Machine Translation, MT evaluation, MT entry generation, NLP tools integration, workflow tool NLP Technology in Documentation and Translation at SAP From the late 80’s early 90’s, SAP has developed and enhanced its own proprietary terminology database (SAPterm) which is fully integrated in their business software product. Currently, version 3 is being used productively. In-house it is the central source for SAP terminology and can be accessed by all SAP employees via a Web interface. The MultiLingual Technology (MLT) group originally started out as a purely Machine Translation group back in 1990. Once its first project to introduce MT for German-to-English as an in-house service for the specialized translators was realized, the group very soon realized that more could be achieved with such technology. The success of this initial project then triggered several others. With the importance of the SAP products in the market growing, the number of customer notes1 (or short: notes) increased accordingly. This posed a problem since the translation had to be available within 24 hours upon release of the source text for the priority 1 notes (Support Level Agreement). These parameters made it impossible to succeed using the conventional methods. In 1994, a cooperation with the MT vendor to implement a workflow tool combining Machine Translation and Translation Memory for the translation of notes was therefore initiated. With this tool, it was possible for 4-6 people to easily translate some 100 to 150 notes (on average 96 words) per day. Once the entire process was set up, thoroughly tested and the technology proved to be stable, the customer notes translation was gradually 1 As one of the world’s largest software houses, SAP AG has a complex system for informing customers about bug fixes in and distributing minor source code corrections for the R/3 System, as well as for distributing additional functionality, best practices and other customer requests. outsourced starting in 1996 (Brundage, McCormick & Pyne 1997). SAP’s popularity in other than the English and German speaking markets, of course, led to the …

Content truncated to 500 words