N-gram을 활용한 특수목적중국어의 어휘다발 연구: AI-Hub 학습용 데이터를 중심으로

Abstract

In this study, as the demand for tourism increases in Korea-China relations, the need for tourism vocabulary learning is increasing when using apps and machine translation. Therefore, in order to construct basic data necessary for learning tourism Chinese vocabulary, the type of N-gram vocabulary was analyzed based on a native Chinese corpus on the AI-Hub platform. As a result of the analysis, in the 2-gram, many high-frequency vocabulary bundles similar to the types appearing in general-purpose Chinese learning contents appeared. The characteristic of the 3-gram vocabulary bundle is that adverbs occupy a large proportion in the tourist Chinese vocabulary list. The characteristic of the 4-gram vocabulary bundle is that it can be used as an independent expression as a lot of professional vocabulary appears and syllables grow. Next, tourism Chinese vocabulary was categorized, classified into verb phrases and noun phrases, and analyzed through frequency of use and example sentences. The verb phrase is a type of 'pronoun + auxiliary verb + verb', and includes 我 想 去, 我 想 买, and 你 能 告诉. In addition, auxiliary verb 能 often appears in dialogues. The noun phrase was categorized into 'investigation + noun' and 'association + noun', and the biggest feature is that vocabulary bundles that include professional vocabulary 飞机, 着陆, 登机, 航班, and 预 settings appear frequently in 4-gram.