網際網路的迅速發展帶給人們方便。但每天都有大量的文本資訊需要閱讀，此時可使用意見探勘，從大量文本中擷取人們感興趣之部分和重要的意見觀點，幫助我們快速地掌握並理解文章中持有者所要陳述的意見觀點及其主張。一般意見句可分為四個部分，包括意見主題、意見持有者、意見主張及意見情感，本研究目標是在於辨識意見持有者，本研究提出類神經網路方法，首先找出文章的意見句，再辨識意見句中的文章作者意見以及意見持有者。
利用自然語言處理之方法辨識文章作者以及意見持有者，其中前處理方法包括斷詞(Tokenize)、蒐集意見詞、還原字根(Stemming)、尋找意見句、詞性標記(POS)、具名實體辨識(NER)和文章作者以及意見持有者之特徵值擷取，本論文利用詞彙相關資訊、詞性相關資訊、標點符號相關資訊、具名實體相關資訊、句法相關資訊、意見詞資訊以及文句組成相關資訊等特徵辨識文章中意見句之文章作者意見以及意見持有者。
實驗成果顯示在英語新聞文章中，文章作者意見辨識可以達到F-1值99.44%的效能；意見持有者辨識可以達到F-1值81.71%的效能。

The rapid development of the Internet has brought convenience to people. However, there is a lot of text information that we need to read every day . At this time, we can use opinion exploration techniques to extract people's interests and important opinions from a large number of texts, helping us quickly grasp and understand the opinion viewpoints and claims of holders of the articles. In general, an opinion sentence can be divided into four parts, including opinion topic, opinion holder, opinion claim and opinion sentiment. The purpose of this study is to identify the opinion holder. This study proposes a neural network method. First we find the opinion sentences of the article, and then identify the author of the article in the opinion sentences and the holder of the opinion.

The method of natural language processing is used to identify the author of the article as well as the opinion holder, in which the method includes tokenization, collecting opinion words, stemming, finding opinions, part-of-speech tagging, recognizing the named entity and the author of the article and the feature extraction. In the feature extraction section, this thesis uses the features of lexical related information, part of speech related information, punctuation related information, named entity related information, syntactic related information, opinion word information and sentence information to identify the article's opinion sentences, author's opinions and opinions holder.
The experimental results show that, the article author's opinion recognition can achieve 99.44% of the F-1 value and the opinion holder extraction can get 81.71% of the F-1 value.

[1] Kim, S. M., & Hovy, E. (2004, August). "Determining the sentiment of opinions." In Proceedings of the 20th international conference on Computational Linguistics (p. 1367).

[2] Choi, Y., Cardie, C., Riloff, E., & Patwardhan, S. (2005, October). "Identifying sources of opinions with conditional random fields and extraction patterns." In Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing (pp. 355-362).

[3] Michael W., Marc S., & Josef R. (2015, September). "Opinion Holder and Target Extraction for Verb-based Opinion Predicates – The Problem is Not Solved" In Proceedings of the 6th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis (WASSA 2015), pages 148–155, Lisboa, Portugal

[4] Steven, B., Hong, Y., Vasileios, H., & Dan, J. (2004, January). "Automatic Extraction of Opinion Propositions and their Holders." In Proceedings of the AAAI spring symposium on exploring attitude and affect in text: theories and applications

[5] Chen, C. R. (2018, July) . " Automatically Extracting Opinion Sentences and Identifying Opinion Holders in News" National Taiwan Normal University , Thesis

[6] EBS美樂顧問中心-英語實力養成、專業能力發揮: http://ebseducation. pixnet.net/blog/post/406370224%E3%80%90%E5%95%86%E7%94%A8email%E3%80%919%E5%80%8B%E5%B8%B8%E7%94%A8%E7%9A%84%E5%8B%95%E8%A9%9E,%E6%9C%89%E6%95%88%E7%8E%87%E7%A2%BA%E5%AF%A6%E8%A1%A8%E9%81%94acknowledge

[7] Exam English Home Page: https://www.examenglish.com/vocabulary/b1_perso nal_feelings.htm

[8] Macmillan Dictionary: http://www.macmillandictionary.com/thesaurus- category/
british/to-give-your-opinion

[9] Thesaurus.com: http://www.thesaurus.com/browse/give%20opinion/

[10] Cortes, C., & Vapnik, V. (1995). "Support-vector networks." Machine learning, 20(3), 273-297.

[11] Lafferty, J., McCallum, A., & Pereira, F. C. (2001). "Conditional random fields: Probabilistic models for segmenting and labeling sequence data." In Proceedings of the eighteenth international conference on machine learning, ICML (Vol. 1, pp. 282-289).

[12] Manning, C. D., Surdeanu, M., Bauer, J., Finkel, J. R., Bethard, S., & McClosky, D. (2014, June). "The Stanford CoreNLP natural language processing toolkit." In ACL (System Demonstrations) (pp. 55-60)

[13] Poter’s Stemmer :https://tartarus.org/martin/PorterStemmer/

[14] Zhang, Y. H. (2006, July). "Combining the Supervised and Unsupervised Approaches to Identifying Opinion Holders in News." National Taiwan Normal University , Thesis

[15] NLTK: https://www.nltk.org/book/

[16] Opinion Lexicon: https://github.com/jeffreybreen/twitter-sentiment-analysis-tutir
ial-201107/tree/master/data/opinion-lexicon-English

[17] 李佳穎、古倫維和陳信希，2009，"意見持有者辨識及其意見立場分析"，國立台灣大學資訊工程所碩士論文。

[18] Embedding Layer: https://www.zhihu.com/question/45027109

[19] CNN: https://www.jishuwen.com/d/2qIn/zh-tw

[20] Kim, Y. "Convolutional Neural Networks for Sentence Classification" (2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1746–1751)
[21] Das, D., & Bandyopadhyay, S. (2011, July). "Emotions on Bengali blog texts: role of holder and topic." In Advances in Social Networks Analysis and Mining (ASONAM), 2011 International Conference on (pp. 587-592). IEEE

[22] Elarnaoty, M., AbdelRahman, S., & Fahmy, A. (2012). "A machine learning approach for opinion holder extraction in Arabic language." arXiv preprint arXiv:1206.1011.

[23] Kim, S. M., & Hovy, E. (2005, July). "Identifying opinion holders for question answering in opinion texts." In Proceedings of AAAI-05 Workshop on Question Answering in Restricted Domains (pp. 1367-1373).

[24] Wiegand, M., & Klakow, D. (2012, April). "Generalization Methods for In-Domain and Cross-Domain Opinion Holder Extraction "

[25] Yang, B., & Cardie, C. (2013, August). "Joint Inference for Fine-grained Opinion Extraction"

[26] Kim, S. M., & Hovy, E. (2006,July). "Extracting Opinions, Opinion Holders, and Topics Expressed in Online News Media Text"