PENS: A Machine-aided English Writing System for Chinese Users

PENS: A Machine-aided English Writing System for Chinese Users Ting Liu1 Ming Zhou Jianfeng Gao Endong Xun Changning Huang Natural Language Computing Group, Microsoft Research China, Microsoft Corporation 5F, Beijing Sigma Center 100080 Beijing, P.R.C. { i-liutin, mingzhou, jfgao, i-edxun, cnhuang@microsoft.com } Abstract Writing English is a big barrier for most Chinese users. To build a computer-aided system that helps Chinese users not only on spelling checking and grammar checking but also on writing in the way of native-English is a challenging task. Although machine translation is widely used for this purpose, how to find an efficient way in which human collaborates with computers remains an open issue. In this paper, based on the comprehensive study of Chinese users requirements, we propose an approach to machine aided English writing system, which consists of two components: 1) a statistical approach to word spelling help, and 2) an information retrieval based approach to intelligent recommendation by providing suggestive example sentences. Both components work together in a unified way, and highly improve the productivity of English writing. We also developed a pilot system, namely PENS (Perfect ENglish System). Preliminary experiments show very promising results. Introduction With the rapid development of the Internet, writing English becomes daily work for computer users all over the world. However, for Chinese users who have significantly different culture and writing style, English writing is a big barrier. Therefore, building a machine-aided English writing system, which helps Chinese users not only on spelling checking and grammar checking but also on writing in the way of native-English, is a very promising task. Statistics shows that almost all Chinese users who need to write in English1 have enough knowledge of English that they can easily tell the difference between two sentences written in Chinese-English and native-English, respectively. Thus, the machine-aided English writing system should act as a consultant that provide various kinds of help whenever necessary, and let users play the major role during writing. These helps include: 1) Spelling help: help users input hard-to-spell words, and check the usage in a certain context simultaneously; 2) Example sentence help: help users refine the writing by providing perfect example sentences. Several machine-aided approaches have been proposed recently. They basically fall into two categories, 1) automatic translation, and 2) translation memory. Both work at the sentence level. While in the former, the translation is not readable even after a lot of manually editing. The latter works like a case-based system, in that, given a sentence, the system retrieve similar sentences from translation example database, the user then translates his sentences by analogy. To build a computer-aided English writing system that helps Chinese users on writing in the way of native-English is a challenging task. Machine translation is widely used for this purpose, but how to find an efficient way in which human collaborates well with computers remains an open issue. Although the quality of fully automatic machine translation at the sentence level is by no means satisfied, it is hopeful to 1 Now …

Content truncated to 500 words