Machine Translation

Walter von Hahn

1. Introduction

This paper presents a short overview of the state of the art in machine translation with some examples from ongoing projects. The second part summarizes challenging topics of research. The last paragraphs share some practical experience on how to organize efficient projects.

2. Definitions

Machine Translation (MT) can be defined as a translation where the initiative is with a computer system, either autonomously (FAHQT = Fully Automatic High Quality Translation) or where the user is asked to apply post-editing or pre-editing, or to answer clarification / disambiguation dialogues.

Machine Aided Translation (MAT) is human translation supported by a computer system. Support is available by lexical data, grammatical help, translation memory, domain information and organizational support.

Translation Types and Time Constraints

Machine translation systems belong to different performance types: The user needs translations within hours for informal papers, such as infos, working documents, notes or letters. Users will expect a translation within days for research papers or commercial reports, for instructions or other technical documents. More formal descriptions like annual reports, proceedings or official company materials should be available within weeks. Only translations for books are scheduled in the degree of years.

These performance types are tied to quality characteristics and acceptance criteria of users: The faster the system has to work, the more the user will accept minor quality.

121