공기 관계 네트워크를 이용한 감정명사의 사용 양상 분석

Abstract

This paper aims to analyze emotion nouns based on co-occurrence relation networks which are constructed from Trends 21 Corpus, a large-scale newspaper corpus. These networks reflect the degrees of co-occurrence relation measured by t-score and show various aspect of the use of Korean emotion nouns during the period of 2000~2008. In particular, we find that co-word of 30 emotion nouns can be grouped uniquely in their networks by the year. This kind of approach to emotion nouns using co-occurrence networks reveals interesting semantic aspects which have not been captured before.