Machine Translation and Multi-Lingual Text Processing

In this paper, I look at some of the problems of Machine Translation (MT) as compared to multi-lingual text processing and text retrieval. In particular, I discuss the question of the evaluation of the output of MT systems, as opposed to the output of other NLP applications. Finally I raise the question of the political implications of the choice of input and output languages for MT systems within the larger context of a world where electronic communication renders access to textual information crucial for the maintenance of less established languages. Machine Translation It is well-known that Machine Translation (MT) is a type of application which presents most of the most difficult tasks of Natural Language Processing (NLP). It involves at the very least the following three components: a. analysis of the source text, b. generation of the target text, and c. mapping between the two representations needed for tasks (a) and (b) The choice of levels of representation for (a) and (b), ..