감정 기반 블로그 문서 분류를 위한 부정어 처리 및 단어 가중치 적용 기법의 효과에 대한 연구
A Study on Negation Handling and Term Weighting Schemes and Their Effects on Mood-based Text Classification
0
조회
0
다운로드
5
인용
0
참고문헌
키워드
초록
일상생활에서 많이 쓰이는 블로그 문서를 분석하는 것은 다양한 웹 응용서비스를 연결할 수 있는 중요한 단초를 제시하므로, 블로그 문서에 담긴 감정을 파악하는 것을 매우 유용한 일이다. 본 논문에서는 블로그 문서에 존재하는 감정을 보다 정확하게 분류하기 위해 부정어 처리와 새로운 단어 가중치의 적용이 성능에 미치는 영향에 대해 탐구한다. 특히, 감정 단서(clue)가 내재된 정규화된 부정어 n-gram을 통해 부정어 처리를 고도화하고 말뭉치기반 단어 가중치 계산법(Corpus-specific Term Weighting, CSTW)을 통해 감정 분류 성능향상을 살펴보기로 한다. 검증을 위해 블로그 문서들로 정답 말뭉치를 구축하고 감정 흐름 분석(Enhanced Mood Flow Analysis, EMFA)과 지지벡터기계기반 감정 분류(Support Vector Machine based Mood Classification, SVMMC)의 두 가지 분류기법에 대해 실험을 하였다. 정규화된 부정어 n-gram의 적용은 EMFA에서 점진적인 감정 분류 성능 향상을 보여주었으며, CSTW의 적용은 TF*IDF나 TF에 비해 보다 높은 감정 분류 성능을 나타내었다.
Abstract
Mood classification of blog text is an interesting problem, with a potential for a variety of services involving the Web. This paper introduces an approach to mood classification enhancements through the normalized negation n-grams which contain mood clues and corpus-specific term weighting (CSTW). We’ve done experiments on blog texts with two different classification methods: Enhanced Mood Flow Analysis (EMFA) and Support Vector Machine based Mood Classification (SVMMC). It proves that the normalized negation n-gram method is quite effective in dealing with negations and gave gradual improvements in mood classification with EMFA. From the selection of CSTW, we noticed that the appropriate weighting scheme is important for supporting adequate levels of mood classification performance because it outperforms the result of TF*IDF and TF.
참고문헌 (14)
- Gilad Mishne, "“Experiments with Mood Classification in Blog Posts”, Style2005 at SIGIR 2005", ~ (2005)
- Yuchul Jung, "“Determining Mood for a Web Blog by Combining Multiple Sources of Evidence”, Proc. of IEEE", WIC, pp.2007, ~
- Kushal Dave,Steve Lawrence and David M.Pennock, "“Mining the Peanut Gallery: Opinion Extraction and Semantic Classification of Product Reviews”, Proc. of the 20th Int. World Wide Web Conf. (WWW-03), pp. 519-528, 2003", ~ (2003)
- Jin-Cheon Na, "Use of negation phrases in automatic sentiment classification of product reviews", Library collections Acquisitions & Technical Services, 29, pp.180~191 (2005)
- Bo Pang, "“Thumbs up? Sentiment Classification using Machine Learning Techniques”, Proc. of the ACL-02 Conf. on Empirical Methods in Natural Language Processing (EMNLP), pp. 79-86, 2002", ~ (2002)
- Bo Pang, "“A Sentiment Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts”, Proc. of the 42nd Annual Meeting on Association for Computational Linguistics (ACL), no (271), 2004", ~ (2004)
- Alistair Kennedy, "Sentiment Classification of Movie Reviews Using Contextual Valence Shifters", Computational Intelligence, 22(2), pp.110~125 (2006)
- Bradley, M. M, "Affective norms for English words (ANEW). Gainesville, FL. The NIMH Center for the Study of Emotion and Attention, University of Florida, 1999", ~ (1999)
- Rada Mihalcea, "“A Corpus-based approach to finding happiness”, Proc. of AAAI Spring Symposium on Computational Approaches to Weblogs, 2006", ~ (2006)
- Chul Min Lee, "Toward Detecting Emotions in Spoken Dialogs", IEEE Transaction on Speech and Audio Processing, 13(2), pp.293~303 (2005)
- Hugo Liu, "ConceptNet-a practical commonsense reasoning tool-kit", BT Technology Journal, pp.211~226 (2004)
- Push Singh, "“Open Mind Common Sense: Knowledge acquisition from the general public”, Proc. of the 1st Int. Conf. on Ontologies, Databases, and Applications of Semantics for Large Scale Information Systems, 2002", ~ (2002)
- Ekman, P, "Facial expression of emotion", American Psychologies, 48, pp.384~392 (1993)
- Mehrabian, A., "Framework for a comprehensive description and measurement of emotional states", Genetic Social and General Psychology Monographs, 121(3), pp.339~361 (1995)