워드임베딩에 기반한 러시아어 감정어휘의미 연구 -슬픔의 감정형용사를 중심으로
Lexical Meaning of Russian Emotion Words based on Word Embedding: Focusing on Adjectives of Sadness
1
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
This study aims to demonstrate the linguistic significance of classifying Russian emotional adjectives based on word embeddings. To achieve this, the study examines the types of emotional vocabulary. Secondly, it reviews previous research on the semantics of Russian emotional vocabulary. Thirdly, it explores semantic associates based on word embeddings. Fourthly, the study groups Russian emotional adjectives using word embedding-based semantic associates and analyzes differences between these results and those of traditional lexical semantic research. Finally, it discusses the potential role of word embeddings in the semantic analysis of emotional vocabulary. This study utilized three models built by the Language Technology Group at the University of Oslo. To evaluate the performance of the models, an analogy evaluation was conducted to assess how well the language models understand the semantic relationships between words. As a result, it was confirmed that a sufficient quantity and quality of data, accurate preprocessing, and effective parameter combinations for learning are necessary to construct high-quality models. Through the analysis of the similarity between "тоскливый" and "грустный", it was examined how the synonyms of these adjectives are represented. Therefore, exploring the semantic relationships of emotional vocabulary through word embeddings can serve as a valuable tool in linguistic research, demonstrating the complementary relationship between linguistic intuition and machine learning results.
Abstract
This study aims to demonstrate the linguistic significance of classifying Russian emotional adjectives based on word embeddings. To achieve this, the study examines the types of emotional vocabulary. Secondly, it reviews previous research on the semantics of Russian emotional vocabulary. Thirdly, it explores semantic associates based on word embeddings. Fourthly, the study groups Russian emotional adjectives using word embedding-based semantic associates and analyzes differences between these results and those of traditional lexical semantic research. Finally, it discusses the potential role of word embeddings in the semantic analysis of emotional vocabulary. This study utilized three models built by the Language Technology Group at the University of Oslo. To evaluate the performance of the models, an analogy evaluation was conducted to assess how well the language models understand the semantic relationships between words. As a result, it was confirmed that a sufficient quantity and quality of data, accurate preprocessing, and effective parameter combinations for learning are necessary to construct high-quality models. Through the analysis of the similarity between "тоскливый" and "грустный", it was examined how the synonyms of these adjectives are represented. Therefore, exploring the semantic relationships of emotional vocabulary through word embeddings can serve as a valuable tool in linguistic research, demonstrating the complementary relationship between linguistic intuition and machine learning results.
참고문헌 (32)
- 김보라, "단어임베딩을 활용한 러시아어 어휘의미 연구", 슬라브연구, pp.183~198 (2021)
- 김호용, "우리말샘 사전을 이용한 단어 의미 유사도 측정모델 개발", 한국콘텐츠학회 종합학술대회 논문집, pp.3~4 (2018)
- 남길임, "한국어 텍스트 감성 분석" (2022), 커뮤니케이션북스
- 동완, "노한사전" (1998), 주류출판사
- 안혁, "러시아어의"슬픔"의미연구 : 감정언어의 인지의미론적 고찰", 슬라브연구, pp.129~159 (2015)
- 유원준, "딥러닝을 이용한 자연어 처리 입문" (2021), WikiDocs
- 이기웅, "러시아와 감정의 토포스" (2016), 경북대학교 출판부
- 이기창, "한국어 임베딩" (2019), 에이콘
- 이민우, "역사 말뭉치 분석을 통한 감정 어휘의 의미 영역 변화 양상 연구", 한국어 의미학, 70, pp.69~100 (2020)
- 임희석, "자연어처리 바이블핵심이론 응용시스템 딥러닝" (2020), 휴먼사이언스
- 정정원, "러시아어 장소 은유 감정 술어 구문 연구", 슬라브어 연구, 14(2), pp.301~318 (2009)
- Apresjan, V., "‘Fear’ and ‘Pity’ in Russian and English from a lexico-graphic perspective", International Journal of Lexicography, 85, pp.111~ (1997)
- Berners-Lee, T., "The semantic web", Scientific American, 284(5), pp.34~43 (2001)
- Chung, D., "A Case Study of Visualizing Emotions with Social Media Emotion Analysis Focused onMedia. Art Cases", Archives of Design Research, 35(1), pp.237~257 (2022)
- Clark, Stephen, "The Handbook of Contemporary Semantic Theory", pp.493~522 (2015), Wiley-Blackwell
- Ekman, P., "Universals and cultural differences in facial expressions of emotion", Nebraska Symposium on Motivation, 19, pp.207~283 (1971)
- Frijda, N. H., "Everyday Conceptions of Emotion. NATO ASI Series, vol 81" (1995), Springer
- Korogodina, Olga, "Evaluation of Vector Transformations for Russian Word2Vec and FastText Embeddings", Proceedings of the 30th International Conference on Computer Graphics and Machine Vision (GraphiCon 2020) (2020)
- Kutuzov A., "Analysis of Images, Social Networks and Texts. AIST 2016. Communications in Computer and Information Science" (2017), Springer
- Mikolov, T., "Efficient estimation of word representations in vector space", Proceedings of Workshop at ICLR (2013)
- Ortony, Andrew, "What's Basic About Basic Emotions?", Psychological review, 9, pp.315~331 (1990)
- Oatley, K., "Towards a Cognitive Theory of Emotions", Cognition and Emotion, 1(1), pp.29~50 (1987)
- Plutchik, Robert, "The Nature of Emotions", American Scientist, 89, pp.3~33 (1980)
- Абрамов, Н, "Словарь русских синонимов и сходных по смыслувыражений —" (1999), Русские словари
- Апресян Ю.Д. и др, "Новый объяснительный словарь синонимов русского языка", ~ (2004)
- Бабенко Л.Г. Лексические, "Средства обозначения эмоций в русском языке", Издательство Уральского Университета, ~ (1989)
- Иорданская Л. Н, "Попытка лексикографического толкования группырусских слов созначением чувства", Машинный перевод и прикладная лингвистика, 13, pp.3~26 (1970)
- "NLTK"[URL]
- "RusVectores"[URL]