재난재해 소셜 미디어 텍스트 감성분석을 위한 키워드 노이즈 제거 방법론 연구
A Methodology of Filtering Irrelevant Keywords for Sentiment Analysis of Disaster-related Social Media Texts
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study proposes a methodology for constructing linguistic resources in order to eliminate irrelevant keywords from social media texts related to disasters such as earthquakes or typhoons. When collecting disaster-related social media texts for sentiment analysis, a large number of noisy keywords metaphorically used such as ‘pupil-earthquake = astonishment’ is observed. In this regard, filtering these linguistic noisy expressions plays a crucial role in performing an accurate text classification or sentiment analysis. In this study, two types of linguistic patterns are examined for filtering noisy expressions in natural & social disaster-related texts and a bootstrap method based on the DECO Korean electronic dictionary and Local-Grammar Graph(LGG) formalism is suggested. In this way, for 6 keywords, around 110~ 470 patterns per keyword are described in LGGs. By applying them to a new corpus through the DECO Noise-Filter platform, we obtained about 88.4% of f-measure. The methodology suggested in this study may be adopted in filtering other types of noisy expressions, which will improve the reliability of the performance of sentiment analysis of social media texts.
Abstract
This study proposes a methodology for constructing linguistic resources in order to eliminate irrelevant keywords from social media texts related to disasters such as earthquakes or typhoons. When collecting disaster-related social media texts for sentiment analysis, a large number of noisy keywords metaphorically used such as ‘pupil-earthquake = astonishment’ is observed. In this regard, filtering these linguistic noisy expressions plays a crucial role in performing an accurate text classification or sentiment analysis. In this study, two types of linguistic patterns are examined for filtering noisy expressions in natural & social disaster-related texts and a bootstrap method based on the DECO Korean electronic dictionary and Local-Grammar Graph(LGG) formalism is suggested. In this way, for 6 keywords, around 110~ 470 patterns per keyword are described in LGGs. By applying them to a new corpus through the DECO Noise-Filter platform, we obtained about 88.4% of f-measure. The methodology suggested in this study may be adopted in filtering other types of noisy expressions, which will improve the reliability of the performance of sentiment analysis of social media texts.
참고문헌 (15)
- 고아라, "조사"같이"와"처럼"의 의미와 기능에 대한 연구", 건지인문학, 9, pp.5~30 (2013)
- 김진해, "은유적 합성명사의 결합관계와 인지언어학적 해석", 國語學, (70), pp.29~57 (2014), 국어학회[DOI]
- 남지순, "코퍼스 분석을 위한 한국어 전자사전 구축방법론" (2018), 도서출판 역락
- 박태연, "재난안전정보의 통합 관리를 위한 분류체계 현황분석 및 개선방안에 관한 연구", 한국비블리아학회지, 28(3), pp.125~150 (2017), 한국비블리아학회[DOI]
- 신봉희, "빅 데이터를 이용한 재해 정보 지원에 관한 연구", 한국융합학회논문지, 9(8), pp.25~32 (2018), 한국융합학회[DOI]
- 신자행, "빅 데이터 기반의 재난정보관리 방안", 석사, pp.2016
- 안길승, "효과적인 산업재해 분석을 위한 텍스트마이닝 기반의 사고 분류 모형과 온톨로지 개발", 한국안전학회지, 32(5), pp.179~185 (2017), 한국안전학회
- 임채훈, "국어 비유구문의 의미 연구 - ‘처럼’, ‘만큼’을 중심으로 -", 한국어 의미학, 10, pp.209~226 (2002), 한국어의미학회
- 유광훈, "DecoTex Users' Manual. DICORA-TR-2017-12", DICORA, 한국외국어대학교, ~ (2017)
- 황창회, "SNS 사용자 생성문에 대한 코퍼스 수집 시스템 소개: Deco Crawlers. DICORA-TR-01-2018", DICORA, 한국외국어대학교, ~ (2018)
- Baek. S., "Disaster Anxiety Measurement and Corpus-Based Content Analysis of Crisis Communication", IEEE International Conference on Systems, Man, and Cybernetics, pp.1789~1794 (2013)
- Gross, M., "Finite-State language processing" (1997), the MIT Press
- Gross, M., "Nouvelles applications des graphes d’automates finis à la description linguistique", Lingvisticae Investigationes, 22(1-2), pp.249~262 (1999)
- Matherson, D., "The performance of publicness in social media: tracing patterns in tweets after a disaster", Media, Culture & Society, 40(4), pp.584~599 (2018)
- Paumier, S., "De la reconnaissance de formes linguistiques a l’analyse syntaxique", Ph.D., pp.2003