跳到主要內容

臺灣博碩士論文加值系統

(44.192.67.10) 您好!臺灣時間:2024/11/09 18:52
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:胡佳男
研究生(外文):Jia-Nan Hu
論文名稱:使用支持向量回歸與規則式於片語層次Valence-Arousal預測
論文名稱(外文):Using SVR and Rules on pharse Valence-Arousal Prediction
指導教授:賴國華禹良治禹良治引用關係
指導教授(外文):Guo-Hua LaiLiang-Zhi Yu
口試委員:楊正仁
口試委員(外文):Zheng-Ren Yang
口試日期:105年07月
學位類別:碩士
校院名稱:元智大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2017
畢業學年度:105
語文別:中文
論文頁數:34
中文關鍵詞:情感計算詞向量SVR修飾結構ValenceArousal
外文關鍵詞:affective computingword2vecSVRmodified structureValenceArousal
相關次數:
  • 被引用被引用:3
  • 點閱點閱:346
  • 評分評分:
  • 下載下載:64
  • 收藏至我的研究室書目清單書目收藏:2
隨著網絡的普及以及社交平臺的迅速發展,越來越多的用戶在網絡上交流觀點,分享心得,發表看法。因此累積了大量的情感資料。情感計算作爲自然語言處理領域的一個熱門研究方向,對於網絡上巨量信息的情感計算也成爲了情感分析的一個重要方向。以往的情感分析研究大部分都是離散型研究,事先將情感分爲不同的類別,這種分析方法存在很多缺陷。近年來也有一種基於維度型的新的情感分析方法,將情感詞的情感強度轉換成兩組連續的分值Valence(表示情感的正負程度)和Arousal(表示情緒的平靜和激動的程度)對其進行量化,然後針對於情感詞進行計算得出文本的情感強度。
本文是基於維度型的情感分析方法的基礎上,以一個包含1653詞的中文Valence-Arousal情感詞典作爲研究基礎,首度引入詞向量加以SVR的訓練模型來對情感詞Valence和Arousal進行預測,並且針對於文本中情感詞修飾結構對文本情感強度的計算問題的缺陷進行改進。以前的基於維度的情感分析方法首先找出句子的情感詞和修飾結構,在修飾結構方面進行類別劃分和給予權重的方式進行計算修飾後情感強度,本文中對於修飾結構,擬對於不同的修飾結構給予不同的修飾規則,最終完成對片語文本情感強度的計算。
With the rapid development of the network and the popularity of social networking platforms, more and more users on the network exchange views, share experiences, express their views. Therefore, there are a large amount of emotional information. Affective computing is a hot research direction in the fileid of natural language processing. Affective computing has also become an important direction of sentiment analysis for huge emotional information on network . In the past,most study put the emotions into different categories, this brings many defects. In recent years, there is also a new emotional dimension type analysis method based on the emotional intensity of the emotional word is converted into two successive scores whichi are Valence (the degree of positive or negative) and Arousal (the degree of excitement and calm), and then calculated for emotional words in the text drawn emotional intensity.
This paper is based on the dimension type of sentiment analysis method, a dictionary that contains 1653 Chinese dictionary Valence-Arousal emotional words as the research base. first introduced the word vector to SVR training model to predict the emotional word’s Valence and Arouse, and emotional words for modification of the defect structure of the text emotional intensity calculation make improvements in the text. Emotional analysis dimension is based on the previous first identify the emotional words and modified structure of the sentence, and put modifier structure into some categories and give weight manner calculated after modification of emotional intensity in the modified structure, herein for modifying the structure, intended for different modified structure given different modification rules, complete the calculation of the final text of emotional intensity.
目錄
第一章 緒論 1
1.1研究背景與意義 1
1.2本文主要研究內容 3
1.3 論文章節安排 4
第二章 國內外研究現狀 6
2.1國外研究現狀 6
2.2國內研究現狀 7
第三章 論文相關背景知識介紹 10
3.1情感詞典和片語修飾語料庫構建 10
3.2訓練詞向量 11
3.3 SVR支持向量回歸 13
第四章 構建基於片語層次的Valence和Arousal預測模型 15
4.1 構建情感詞VA值預測模型 15
4.2片語資料修飾語處理 18
第五章 實驗結果和分析 21
5.1實驗資料集 21
5.2實驗評價指標選取 25
5.2實驗結果對比 26
5.3實驗誤差分析 28
第六章 總結 29
6.1論文的主要工作和貢獻 29

6.2今後的研究工作 29
參考文獻 31


表目錄
表 1 部分中文情感詞預測結果 17
表 2 部分中文情感詞預測結果 18
表 3 Semeval2016英文片語預料修飾語 23
表 4 中文片語預料修飾語 24
表 5 8個中文預料庫相關情况介紹 25
表 6 爲預料收集過程和人工標記的相關情况整理 26
表 7 英文資料集上不同修飾語的實驗結果 28
表 8 英文資料集上不同團隊之間實驗結果對比 28
表 9 中文文資料集上不同回歸方法實驗結果 29
表 10 中文資料集上使用不同的詞向量的實驗結果 29


圖目錄
圖 1 系統框架圖 8
圖 2 片語情感強度VA預測 11
圖 3 部分中文情感詞預測結果 13
[1] Picard R W. Affective computing[M]. Cambridge: MIT press, 1997.
[2] Kiritchenko S, Mohammad S M. Capturing reliable fine-grained sentiment associations by crowdsourcing and best–worst scaling[C]//Proceedings of The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), San Diego, California. 2016.
[3] Pang B, and Lee L. Opinion mining and sentiment analysis[J]. Foundations and Trends in Information Retrieval, 2008, 2(1-2): 1-135
[4] Bradley M M, Lang P J. Affective Norms for English words (ANEW): Instruction manual and affective ratings[R]. Technical report C-1, the center for research in psychophysiology, University of Florida, 1999.
[5] Mikolov T, Sutskever I, Chen K, et al. Distributed representations of words and phrases and their compositionality[C]//Advances in neural information processing systems. 2013: 3111-3119.
[6] Mikolov T, Chen K, Corrado G, et al. Efficient estimation of word representations in vector space[J]. arXiv preprint arXiv:1301.3781, 2013.
[7] Astudillo R F, Amir S, Ling W, et al. Inesc-id: A regression model for large scale twitter sentiment lexicon induction[J]. SemEval-2015, 2015: 613.
[8] Warriner A B, Kuperman V, Brysbaert M. Norms of valence, arousal, and dominance for 13,915 English lemmas[J]. Behavior research methods, 2013, 45(4): 1191-1207.
[9] Yogatama D, Faruqui M, Dyer C, et al. Learning word representations with hierarchical sparse coding[C]//Proc. of ICML. 2015.
[10] Faruqui M, Dodge J, Jauhar S K, et al. Retrofitting word vectors to semantic lexicons[J]. arXiv preprint arXiv:1411.4166, 2014.
[11] Taboada M, Brooke J, Tofiloski M, et al. Lexicon-based methods for sentiment analysis[J]. Computational linguistics, 2011, 37(2): 267-307.
[12] Goldberg Y, Levy O. word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method[J]. arXiv preprint arXiv:1402.3722, 2014.
[13] Tang D, Wei F, Qin B, et al. Building Large-Scale Twitter-Specific Sentiment Lexicon: A Representation Learning Approach[C]//COLING. 2014: 172-182.
[14] Kiritchenko S, Mohammad S M. Capturing reliable fine-grained sentiment associations by crowdsourcing and best–worst scaling[C]//Proceedings of The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), San Diego, California. 2016.
[15] Kiritchenko S, Mohammad S M. The effect of negators, modals, and degree adverbs on sentiment composition[C]//Proceedings of the Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis (WASSA). 2016.
[16] 藺璜,郭姝慧. 程度副詞的特點範圍與分類[J]. 山西大學學報(哲學社會科學版). 2003(02)
[17] Mikolov T. Statistical language models based on neural networks[J]. Presentation at Google, Mountain View, 2nd April, 2012.
[18] Turian J, Ratinov L, Bengio Y. Word representations: a simple and general method for semi-supervised learning[C]//Proceedings of the 48th annual meeting of the association for computational linguistics. Association for Computational Linguistics, 2010: 384-394.
[19] Lin J C, Wu C H, Wei W L. Error weighted semi-coupled hidden Markov model for audio-visual emotion recognition[J]. IEEE Transactions on Multimedia, 2012, 14(1): 142-156.
[20] Bengio Y, Ducharme R, Vincent P, et al. A neural probabilistic language model[J]. journal of machine learning research, 2003, 3(Feb): 1137-1155.
[21] Huang E H, Socher R, Manning C D, et al. Improving word representations via global context and multiple word prototypes[C]//Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1. Association for Computational Linguistics, 2012: 873-882.
[22] Bian J, Gao B, Liu T Y. Knowledge-powered deep learning for word embedding[C]//Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer Berlin Heidelberg, 2014: 132-148.
[23] Faruqui M, Dyer C. Improving vector space word representations using multilingual correlation[C]. Association for Computational Linguistics, 2014.
[24] Xu C, Bai Y, Bian J, et al. Rc-net: A general framework for incorporating knowledge into word representations[C]//Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management. ACM, 2014: 1219-1228.
[25] Yu M, Dredze M. Improving Lexical Embeddings with Semantic Knowledge[C]//ACL (2). 2014: 545-550.
[26] Botha J A, Blunsom P. Compositional Morphology for Word Representations and Language Modelling[C]//ICML. 2014: 1899-1907.
[27] Levy O, Goldberg Y. Dependency-Based Word Embeddings[C]//ACL (2). 2014: 302-308.
[28] Luong T, Socher R, Manning C D. Better Word Representations with Recursive Neural Networks for Morphology[C]//CoNLL. 2013: 104-113.
[29] Pennington J, Socher R, Manning C D. Glove: Global Vectors for Word Representation[C]//EMNLP. 2014, 14: 1532-43.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊