Automatic extraction of Chinese multiword expressions with a statistical tool

Abstract

In this paper, we report on our experi-ment to extract Chinese multiword ex-pressions from corpus resources as part of a larger research effort to improve a machine translation (MT) system. For ex-isting MT systems, the issue of multi-word expression (MWE) identification and accurate interpretation from source to target language remains an unsolved problem. Our initial test on the Chinese-to-English translation functions of Systran and CCID’s Huan-Yu-Tong MT systems reveal that, where MWEs are in-volved, MT tools suffer in terms of both comprehensibility and adequacy of the translated texts. For MT systems to be-come of further practical use, they need to be enhanced with MWE processing capability. As part of our study towards this goal, we test and evaluate a statistical tool, which was developed for English, for identifying and extracting Chinese MWEs. In our evaluation, the tool achieved precisions ranging from 61.16% to 93.96 % for different types of MWEs. Such results demonstrate that it is feasi-ble to automatically identify many Chi-nese MWEs using our tool, although it needs further improvement. 1