과학 전문 용어 부가 정의문의 정보 유형 및 패턴 - 『표준국어대사전』 물리·생물 분야 뜻풀이 말뭉치를 대상으로 -

Abstract

Acknowledging the importance of terminological definitions, this study sets out to investigate the categories and types of definition, the strategies and the linguistic markers used in definition in order to analyze and describe the definitions of scientific terms in the Standard Korean Language Dictionary, and more precisely the ‘supplementary definitions’, that is, the second definition following the primary definition in a dictionary entry. This study consists of two main steps. First, information types as appearing in the supplementary definitions are inductively identified based on previous research on definitions’ information types. Second, formal patterns that appear recurrently according to each information type are analysed. The corpus used to retrieve the information types in scientific term definitions and the definition patterns by information types is composed of the supplementary definitions of terminological headwords from the Standard Korean Language Dictionary, and more precisely of the nouns and noun equivalents in the fields of physics and biology. The information types of these supplementary definitions can be classified into some categories: ‘function/action’, ‘origin’, ‘example/hyponym’, ‘shape’, ‘organization’, ‘symbol’, ‘distribution’, ‘hyperonym’, and so on. The extraction of formal patterns by information types was performed by a tool called KNUgram, which is an n-gram-based language analysis tool specifically made for the morphological analysis of Korean language corpora and which uses the concept of skip-gram for the retrieval of consecutive word combinations and discontinuous word combinations. As the information types established for this study do not necessarily show a high frequency or reflect particular formal patterns, it can be said that the establishment of these information types can seemed to be somewhat arbitrary and in that sense shows some limitations. Nonetheless, the value of this study lies in that it seeks to overcome the limitations shown in previous research, including the overly comprehensive classification of conceptual terms and the consequent difficulty to apply it to practical tasks. Moreover, from an educational perspective, it is believed that not only the typical patterns of formal definitions, such as those appearing in specialized textbook corpora, but also patterns of semi-formal and informal definitions can be automatically retrieved and presented to learners of particular fields, thereby contributing to the conceptual understanding of specialized terms.