언어 정보의 계량화와 시각화: 감성 어휘의 분포 의미 표현

초록

The purpose of this article is to examine the whole process of a methodology of quantifying and visualising linguistic information as a case study of sentiment lexicon building. Sentiment lexicons are one of the core language resources in sentiment analysis which aims to detect and classify the sentiments based on semantic concepts similar to human sensibility as an application of language analysis. More recently, the application of linguistic information is going from proposing the quantification of lexical information of frequency to developing a language model using machine learning methods, so-called deep learning, and inferencing linguistic categories. Thus, this study shows word embedding and its visualization for expressing lexical information used for sentiment analysis.

키워드