신문 빅데이터 기반의 언어 계량과 시각화
Quantification of Linguistic Information and its Visualization through an Analysis of Large-Scale Data Collected from Newspaper
0
조회
0
다운로드
11
인용
0
참고문헌
키워드
초록
This paper quantifies information on the frequency of use or distribution of language and examines methods and cases of visually illustrating its results. First, we discuss problems of large-scale language data related to the quantification of language information. Moreover, we look through several methods of quantifying language information and effectively visualizing its results. Through the Dong-A Ilbo newspaper corpus, we have proven that a rather simple linear graph can represent the change of word frequency effectively. Considering the change of frequency according to time, we adopt the statistical measure t-score. We calculate the key words and visualize its results via a word cloud. Also, we process network visualization using Pajek to extract co-occurring words that share contexts with particular words and spot their relevance. Finally, it is shown that a bar graph and multidimensional scaling can be properly utilized to compare the words’ frequency and visualize the similarity.
Abstract
This paper quantifies information on the frequency of use or distribution of language and examines methods and cases of visually illustrating its results. First, we discuss problems of large-scale language data related to the quantification of language information. Moreover, we look through several methods of quantifying language information and effectively visualizing its results. Through the Dong-A Ilbo newspaper corpus, we have proven that a rather simple linear graph can represent the change of word frequency effectively. Considering the change of frequency according to time, we adopt the statistical measure t-score. We calculate the key words and visualize its results via a word cloud. Also, we process network visualization using Pajek to extract co-occurring words that share contexts with particular words and spot their relevance. Finally, it is shown that a bar graph and multidimensional scaling can be properly utilized to compare the words’ frequency and visualize the similarity.
참고문헌 (12)
- 강범모, "한국어 형태소 및 어휘 사용 빈도의 분석 2" (2004), 고려대학교 민족문화연구원
- 강범모, "한국어 사용 빈도" (2009), 한국문화사
- "고려대학교 민족문화연구원 디지털인문학센터 웹 기반 코퍼스 분석 도구"[URL]
- 김일환, "공기 관계 네트워크를 이용한 감정명사의 사용 양상 분석", 한국어학, 49, pp.119~148 (2010), 한국어학회
- 김일환, "대규모 신문 기사의 자동 키워드 추출과 분석 -t-점수를 이용하여-", 한국어학, 53, pp.145~194 (2011), 한국어학회
- 김일환, "저자 판별을 위한 전산 문체론 - 초기 현대소설을 대상으로", 국어국문학, (170), pp.207~240 (2015), 국어국문학회[DOI]
- 김일환, "‘물결21’ 코퍼스의 구축과 활용" (2013), 소명
- 김일환, "디지털 시대 인문학의 미래" (2017), 푸른역사
- 최재웅, "물결21 코퍼스: 공개 웹 자원 및 활용 도구", 민족문화연구, (64), pp.3~23 (2014), 민족문화연구원[DOI]
- Erez A., "Uncharted: Big Data as a Lens on Human Culture" (2013), Riverhead Books
- Zipf, G. K., "The Psychobiology of Language" (1935), Houghton-Mifflin
- Mattew K. G., "Debates in the Digital Humanities" (2012), University of Minnnesota Press