법률 도메인 특화를 위한 중한 기계번역 오류의 귀납적 분석: 중국 민법전의 파파고 번역을 중심으로

The Civil Code is China’s primary legislation and the first law designated as the “code” within China’s legal system. This study selected 1,258 articles from the Chinese Civil Code to create a machine translation system specialized for the Chinese-to-Korean legal domain. First, the selected articles were translated using the Naver Papago machine translation system. After the translation, errors found in each article were systematically compiled and categorized by type, including message, language and grammar, structure. The analysis showed a significant frequency of errors, especially in legal terminology and punctuation usage. Although the study is limited by insufficient data for thorough analysis constructing an extensive corpus of legal texts and scrutinizing their characteristics for a highly specialized machine translation in the legal domain would result in a more apparent domain-specific effect. The changing landscape of the legal market, resulting from factors like legal market liberalization, emphasizes the need for specialized machine translation within the legal domain to keep up with significant demand for legal translation. Considering the situation, this study identifies specific error instances from the perspective of translation research, enabling engineering scholars to recognize the prevalent errors in Chinese-to-Korean legal machine translation, thereby reducing the gap between the roles played in research by translation and engineering scholars and identifying areas for cooperation and collaboration between the two studies.