Chunk Extraction in Business English Correspondences

At present, no scholars have conducted MT-oriented research on the chunk extraction of human–computer interaction in business English correspondences. This chapter proposes to conduct a study on the extraction of business English corresponding chunks through three modules: text pre-processing, automatic extraction and manual processing, focusing on the automatic computer extraction of two to nine words according to the definition and definition criteria of MT-oriented business English corresponding chunks. With the corpus chunk extraction software, the computerized automatic extraction of chunks from the English-Chinese parallel business correspondence corpus is carried out, and then the chunks that do not meet the definition manually are removed to obtain the table of MT-oriented business English letter chunks. This study probes into a key technology for an MT-oriented English-Chinese business letter corresponding chunk bank.