Responsive image
博碩士論文 etd-1130115-184224 詳細資訊
Title page for etd-1130115-184224
論文名稱
Title
於中文臉書文章中考慮事件與地點之負面情緒偵測法和應用
Negative Emotion Detection with Consideration of Events and Places for Chinese Posts on Facebook and Applications
系所名稱
Department
畢業學年期
Year, semester
語文別
Language
學位類別
Degree
頁數
Number of pages
88
研究生
Author
指導教授
Advisor
召集委員
Convenor
口試委員
Advisory Committee
口試日期
Date of Exam
2015-09-11
繳交日期
Date of Submission
2015-12-31
關鍵字
Keywords
負面情緒偵測、資料探勘、情緒分類
Negative emotion detection, Data mining, Emotion classification
統計
Statistics
本論文已被瀏覽 5752 次,被下載 501
The thesis/dissertation has been browsed 5752 times, has been downloaded 501 times.
中文摘要
隨著近幾年社群網站的蓬勃發展,許多學者開始研究社群網站內容中潛在的訊息。在本篇論文中,我們的目標為正確的分類社群網站(臉書)其中貼文的情緒,並從中找出負面的結果來發掘和負面情緒的相關資訊,像是文章屬於哪種負面型態(生氣,傷心,憂慮或是其他),導致負面情緒的事件和作者提到的地點或他的所在地。傳統的情緒分析方法主要可分成兩大類,其一是以情緒詞典為基礎做分類(lexicon based methods) 和機器學習演算法(machine learning based methods)。而本篇論文所使用的方法是基於情緒詞典去做情緒分類,並考慮到文章中的字組關係和事件去做分類。除此之外,在本篇論文的整個情緒分類流程中,也包含了在文章中地點詞的探勘,在無情緒字的文章中的情緒分類法,負面程度判斷等等來幫助我們做情緒判斷和得到許多和負面情緒結果的相關資訊。在我們的實驗中,我們蒐集了許多臉書的文章(12256篇)並拿部分去做訓練(9820篇),剩下的拿來做情緒分類的測試。實驗結果顯示我們提出的情緒分類法相較有名的方法SVM和Naïve Bayesian分類的還要好些(20%與12%),另外還能擷取到負面文章中所提到的事件和地點。因此,本篇論文的方法不但能找出負面文章,還能找到引起作者負面的事件和相關的地點。除此之外,更能應用於大眾對近期熱門事件的情緒分析與個人化的情緒分類等不同議題之上。
Abstract
As the social network gains more popular, people would like to find the related information from social nets to figure out some signs before or after an incident occurred. In this thesis, we would like to classify the emotion of Chinese articles correctly and find the negative emotion. And we even aim to find out event cause and place that are extracted from Facebook or article content. Based on the lexicon based method we consider some factors like word relationship and event which effect the emotion in the article to propose a negative emotion classification rule, a normal sentence classification method and a negative emotion degree calculation scheme to support emotion classification and find out the degree of negative emotion.
In the experiment, many posts from Facebook are extracted as test data and training data to verify accuracy of the proposed methods. Experimental results show that the proposed methods perform better compared to the traditional methods SVM and Naïve Bayesian (20% and 12%). In addition, the proposed methods also extract the events and places related to the posts. Then we apply the methods to two real cases, public emotion about an arbitrary event and personalization. Hence, the proposed methods are not only able to extract the posts with negative emotions, but also able to find out the cause of the events and related place in the posts.
目次 Table of Contents
論文審定書 i
摘要 iii
Abstract iv
List of Figures vi
List of Tables ix

Chapter 1. Introduction 1
Chapter 2. Related Work 8
2.1 Parsing Sentence and Word Extraction 8
2.2 Lexicon Based Methods 10
2.3 Machine Learning Based Methods 12
2.4 Emotion Cause Extraction 15
Chapter 3. Proposed Method 16
3.1 Framework of the Proposed System 16
3.2 Emotion Classification Dictionary 21
3.3 Emotion Classification Methods 24
3.3.1 Emotion Weight Function 24
3.3.2 Composing Word Method 29
3.3.3 Related Place Extraction 35
3.3.4 Negative Emotion Classification Rule 37
3.3.5 Normal Sentence Classification Method 46
3.3.6 Negative Emotion Degree Calculation 51
Chapter 4. Experiments 56
4.1 Evaluation Criteria 56
4.2 Result of Experiment 58
4.3 Applications 69
4.3.1 Public Emotion about the dust explosion at the Formosa Fun Coast water park 69
4.3.2 Personalization 71
Chapter 5. Conclusion and Future Work 74
References 75
參考文獻 References
[1]. S.-C. Chou, W.-T. Qu, Y.-Z. Shi, X.-Z. Shi, and Y.-C. Sun, "Overview on Sentiment Analysis of Chinese Microblogging," Computer Applications and Software, Vol. 30, No. 3, pp. 161-164, 2013.
[2]. C.-C. Huang, M. Eskenazi, J. Carbonell, L.-W. Ku, and P.-C. Yang, "Cross-Lingual Information to the Rescue in Keyword Extraction," Proc. of the 52nd Annual Meeting of the Association for Computational Linguistics, System Demonstration (ACL 2014, System Demonstration), pp. 1–6, 2014.
[3]. H.-H. Huang and H.-H. Chen, "An Annotation System for Development of Chinese Discourse Corpus," Proc. of the 24th Int. Conf. on Computational Linguistics (COLING 2012), pp. 223-230, 2012.
[4]. L.-W. Ku and C.-W. Sun, "Calculating Emotional Score of Words for User Emotion Detection in Messenger Logs," Proc. of the 13th IEEE Int. Conf. on Information Reuse and Integration: Workshop on Empirical Methods for Recognizing Inference in Text II (EM-RITE), pp. 138-143, 2012.
[5]. J. Lei, Y. Rao, Q. Li, X. Quan, and L. Wenyin, "Towards Building a Social Emotion Detection System for Online News," Future Generation Computer Systems, Vol. 37, pp. 438-448, 2014.
[6]. C.-R. Li, C.-H. Yu, and H.-H. Chen, "Predicting the Semantic Orientation of Terms in E-HowNet," Computational Linguistics & Chinese Language Processing, Vol. 17, No. 2, pp. 21-36, 2012.
[7]. W. Li and H. Xu, "Text-Based Emotion Classification using Emotion Cause Extraction," Expert Systems with Applications, Vol. 41, No. 4, pp. 1742-1749, 2014.
[8]. L. Lin, J. Li, R. Zhang, W. Yu, and C. Sun, "Opinion Mining and Sentiment Analysis in Social Networks: A Retweeting Structure-Aware Approach," Proc. of the 2014 IEEE/ACM 7th Int. Conf. on Utility and Cloud Computing (UCC), , pp. 890-895, 2014.
[9]. W.-Y. Ma and K.-J. Chen, "A Bottom-up Merging Algorithm for Chinese UnknownWord Extraction," Proc. of the 2nd SIGHAN Workshop on Chinese Language Processing-Volume 17: Association for Computational Linguistics, pp. 31-38, 2003.
[10]. T.M. Song, J. Song, J.-Y. An, L.L. Hayman, and J.-M. Woo, "Psychological and Social Factors Affecting Internet Searches on Suicide in Korea: A Big Data Analysis of Google Search Trends," Yonsei Medical Journal, Vol. 55, No. 1, pp. 254-263, 2014.
[11]. P.-M. Su and W.-H. Lu, "Predicting Web User’s Tendency of Depression Using Event-Driven Negative Emotion Model," National Cheng-Kung University Master Thesis of Science in Institute of Medical Informatics, 2012
[12]. Y.T. Sun, C.L. Chen, C.-C. Liu, C.-L. Liu, and V.-W. Soo, "Sentiment Classification of Short Chinese Sentences," Proc. of the 22nd Conf. on Computational Linguistics and Speech Processing, pp. 184-198, 2010.
[13]. S. Tan, Y. Li, H. Sun, Z. Guan, X. Yan, J. Bu, C. Chen, and X. He, "Interpreting the Public Sentiment Variations on Twitter," IEEE Transactions on Knowledge and Data Engineering, Vol. 26, No. 5, pp. 1158-1170, 2014.
[14]. F. Tian, P. Gao, L. Li, W. Zhang, H. Liang, Y. Qian, and R. Zhao, "Recognizing and Regulating e-learners’ Emotions Based on Interactive Chinese Texts in e-learning Systems," Knowledge-Based Systems, Vol. 55, pp. 148-164, 2014.
[15]. S.M. Virk, Y.-H. Lee, and L.-W. Ku, "Sinica Semantic Parser for ESWC’14 Concept-Level Semantic Analysis Challenge," The 11th European Semantic Web Conference 2014 (ESWC 2014): Springer, pp. 48-52, 2014.
[16]. J.-H. Wang and C.-C. Lee, "Unsupervised Opinion Phrase Extraction and Rating in Chinese Blog Posts," The 3rd IEEE Int. Conf. on Social Computing (SocialCom 2011), pp. 820-823, 2011.
[17]. J.-H. Wang and T.-W. Ye, "Microblog Sentiment Analysis based on Opinion Target Modifying Relations," Proc. of the 5th Int. Conf. on Social Informatics (SocInfo 2013), pp. 255-267, 2013.
[18]. S. Wen and X. Wan, "Emotion Classification in Microblog Texts Using Class Sequential Rules," Proc. of the 28th AAAI Conf. on Artificial Intelligence, pp. 187-193, 2014.
[19]. C.-H. Wu, Z.-J. Chuang, and Y.-C. Lin, "Emotion Recognition from Text Using Semantic Labels and Separable Mixture Models," ACM Transactions on Asian Language Information Processing (TALIP), Vol. 5, No. 2, pp. 165-183, 2006.
[20]. J.-S. Wu, E.-F. Kao, and C.-N. Lee, "Discovering Hidden Connections among Diseases, Genes and Drugs Based on Microarray Expression Profiles with Negative-Term Filtering," PLoS ONE, Vol. 9, No. 6, pp. 1-15, 2014.
[21]. W. Wu, B. Zhang, and M. Ostendorf, "Automatic Generation of Personalized Annotation Tags for Twitter Users," Proc. of the 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics: Association for Computational Linguistics, pp. 689-692, 2010.
[22]. J. Yan, D.B. Bracewell, F. Ren, and S. Kuroiwa, "The Creation of a Chinese Emotion Ontology Based on HowNet," Engineering Letters, Vol. 16, No. 1, pp. 166-171, 2008.
[23]. C. Yang, K.H. Lin, and H.-H. Chen, "Emotion Classification Using Web Blog Corpora," Proc. of the 2007 IEEE/WIC/ACM Int. Conf. on Web Intelligence: IEEE, pp. 275-278, 2007.
[24]. C.-H. Yu and H.-H. Chen, "Chinese Web Scale Linguistic Datasets and Toolkit," Proc. of the 24th Int. Conf. on Computational Linguistics(COLING), pp. 501-508, 2012.
[25]. B. Yuan, Y. Liu, and H. Li, "Sentiment Classification in Chinese Microblogs: Lexicon-based and Learning-based Approaches," International Proceedings of Economics Development and Research (IPEDR), Vol. 68, No. 1, pp. 1-5, 2013.
[26]. J. Zhao, L. Dong, J. Wu, and K. Xu, "MoodLens: An Emoticon-Based Sentiment Analysis System for Chinese Tweets," Proc. of the 18th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pp. 1528-1531, 2012.
[27]. 中央研究院中文斷詞系統. http://ckipsvr.iis.sinica.edu.tw/. 2010/06/25.
電子全文 Fulltext
本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
論文使用權限 Thesis access permission:自定論文開放時間 user define
開放時間 Available:
校內 Campus: 已公開 available
校外 Off-campus: 已公開 available


紙本論文 Printed copies
紙本論文的公開資訊在102學年度以後相對較為完整。如果需要查詢101學年度以前的紙本論文公開資訊,請聯繫圖資處紙本論文服務櫃台。如有不便之處敬請見諒。
開放時間 available 已公開 available

QR Code