자연언어처리를 기반으로 한한·중 대조 연구 방법에 대한 試探

Abstract

This paper proposes a method of contrastive Korean and Chinese based on Natural Language Processing(NLP). There is a rising interest in studying information processing with the development of science and technology. For this reason, there has been a visible change in language and linguistics research. For example, we have often used corpus, information retrieval, machine translation, and voice recognition systems etc. Furthermore, a lot of people use this for studies first or second language. So we began an exploration for new research based on NLP. There are many problems in Chinese NLP systems, because the Chinese words have various meanings and functions, the computer system keeps generating errors. We need to have a new way of contrasting Korean and Chinese sentence patterns. The first way to contrast them is to use the part of speech(POS) and feature taggers. It is very useful gadgets for helping us to understand what the similarities and differences between word and word are. This is merely a first step for improving the accuracy and efficiency of Computational Language Systems. The tagging data make us prepare for a Computational Language Systems that we able to understand Korean and Chinese sentence patterns. And we propose to collect the data of POS and features taggers. It’s very important material for researchers to analyze the relationship between Korean and Chinese. This subject must be researched, but it's still a work in progress, and we need a good collaborative research project on contrasting of Korean and Chinese.