자질기반 감성분석을 위한 다단어표현 사전 DecoMWE 연구
Constructing Korean Multi-Word Expression Dictionary DecoMWE for Feature-Based Sentiment Analysis,
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
This study proposes Korean Multi-Word dictionary DecoMWE that will be crucial for a reliable feature-based sentiment analysis system. In this study, cosmetic review texts have been analyzed and 4 types of MWEs are classified: two types of Polarity MWEs(Pol-MWE), i.e. General MWEs (GMWE) & Domain-dependent MWEs (DMWE) and two types of Topic MWEs (Top- MWE), i.e. Named-Entity MWEs (EMWE) & Feature MWEs (FMWE). They are meticulously examined and formalized through Local Grammar Graphs (LGGs), technically equivalent to a finite-state transducer. The current DECO- MWE allows to recognize 4 types of MWEs in cosmetic domain and then to annotate them with accurate semantic tags via UNITEX corpus- processing platform. The evaluation of this result reveals about 80% of accuracy, which is promising for a practical use in proceeding feature-based sentiment analysis.
Abstract
This study proposes Korean Multi-Word dictionary DecoMWE that will be crucial for a reliable feature-based sentiment analysis system. In this study, cosmetic review texts have been analyzed and 4 types of MWEs are classified: two types of Polarity MWEs(Pol-MWE), i.e. General MWEs (GMWE) & Domain-dependent MWEs (DMWE) and two types of Topic MWEs (Top- MWE), i.e. Named-Entity MWEs (EMWE) & Feature MWEs (FMWE). They are meticulously examined and formalized through Local Grammar Graphs (LGGs), technically equivalent to a finite-state transducer. The current DECO- MWE allows to recognize 4 types of MWEs in cosmetic domain and then to annotate them with accurate semantic tags via UNITEX corpus- processing platform. The evaluation of this result reveals about 80% of accuracy, which is promising for a practical use in proceeding feature-based sentiment analysis.
참고문헌 (18)
- 권경일, "감정표현 관용구의 의미적 특성에 관한 연구", 문법 교육, (11), pp.119~140 (2009), 한국문법교육학회
- 남지순, "코퍼스 분석을 위한 한국어 전자사전 구축방법론" (2018), 역락
- 유광훈, "SNS텍스트의 오피니언 마이닝을 위한 언어자원 기반 감성 분석 플랫폼 연구", 언어와 정보 사회, 32, pp.123~168 (2017), 언어정보연구소[DOI]
- 정연주, "감정 표현 관용구의 범위 설정", 한국언어문학, (77), pp.31~58 (2011), 한국언어문학회
- 최석재, "구조에서 나타나는 감정 표현 관용구의 의미", 한국어 의미학, 35, pp.311~334 (2011), 한국어의미학회
- 최석재, "어휘의 부류와 감정 표현 관용구의 의미", 한국어학, 55, pp.367~395 (2012), 한국어학회[DOI]
- Baldwin, T, "Handbook of Natural Language Processing", pp.2010, Chapman&Hall
- De Marneffe, M, "Multiword expressions in textual inference: Much ado about nothing", Proceedings of the 2009Workshop on Applied Textual Inference, pp.1~9 (2009)
- Erman, B, "The idiom principle and the open choice principle", Text, 20(1), pp.29~62 (2000)
- Gross, M, "Finite-State Language Processing", pp.329~352 (1997), The MIT Press
- Gross, M, "Analyse Lexicale et Syntaxique: Le Système INTEX, Lingvisticae Investigationes Tome XXII(Volume Spécial)" (1999), John Benjamins Publishing Company
- Kim, K, "(An) Analysis on International Competitiveness of Korean Cosmetics Industry by Diamond Model", M.S, pp.2017
- Lee, K, "Extracting multiword sentiment expressions by using a domain-specific corpus and a seed lexicon", Computational Linguistics, 37(2), pp.267~307 (2011)
- Liu, B, "Sentiment Analysis : Mining Opinions, Sentiments, and Emotion" (2015), Cambridge University Press
- Nam, J, "Korean electronic dictionary DECO TR2015-02" (2015), Digital language and knowledge Contents Research Association (DICORA), Hankuk University of Foreign Studies
- Piao, S, "Extracting multiword expressions with a semantic tagger", Proceedings of the ACL’03Workshop on Multiword Expressions: Analysis, Acquisition and Treatment, pp.49~56 (2003)
- Paumier, S, "De la Reconnaissance de Formes Linguistiques à l’analyse Syntaxique", Ph.D, pp.2003
- Taboada, M, "Lexicon-based methods for sentiment analysis", Computational Linguistics, 37(2), pp.267~307 (2011)