The Processing of English Compound and Complex Words in an English-Chinese Machine Translation System

In a machine translation system the information of the words of the source language should be available before any translation process can begin. The information of simple words can be obtained only by entering a word with all its relevant information into the lexicon. On the other hand, compound words and complex words, it seems, can be handled in a satisfactory way by lexical redundancy rules, and will thus help keep down the size of the lexicon. This paper argues that lexical redundancy rules are not as useful as they may seem to be for a machine translation system, and both their limitations and functions will be examined in depth. In addition, detailed discussions on the various problems that may arise during analyzing and translating of compound and complex words are presented.