國民在資訊科技的發展下，使用者自建內容隨之大量產生，其內容可以是對經濟、社會或政治議題發帖討論，投票響應。透過文字探勘技術，以使用者自建內容取代民意調查的原始資料，用以反映社會輿論或民意動向已是可思考的方向；本篇論文中提出一個議題分析架構，藉由自然語言結構建構出議題的討論面向，並以Stanford Parser搭配NTUSD情感字典實作情感分析流程，以期能彈性的檢視臺灣網民對於兩岸服務貿易協定議題的立場資訊以及情感表達。為了實作議題分析架構，我們蒐集網民於BBS網站PTT政治討論相關版面的討論內容，約20萬篇文章內容並進行過濾，最後採用其中15,000篇文章，藉由與相關文獻的對照，發現網民的整體情緒為負向態度，以及議題的不同面向之討論程度並不平均，著重討論的面向為議題的間接面向。

User generated content (UGC) is increasing quickly due to the advance of information technology. A post about enterprise finance condition or social political issue in forums is part of UGC. Nowadays applying text mining and sentiment analysis techniques on UGC has been shown to be a promising alternative to the traditional polls. In this work, we propose a framework for discovering opinions on issues discussed on forums, specifically those related to Cross-Strait Agreement on Trade in Services. In particular, we identify issues by finding frequent nouns and then discover the pertaining opinions on the posts. Over Fifteen thousand posts on PTT forums are processed using the proposed approach. Finally, we discover that the overall emotion on various issues tend to be negative, whereas the amount of discussion and the degree of negative emotions vary dramatically.

論文審定書 i
中文摘要 ii
英文摘要 iii
第一章 、緒論 1
第一節 、研究背景 1
第二節 、研究動機 2
第三節 、研究問題 3
第二章 、文獻探討 6
第一節 、社會議題與立場 6
第二節 、情緒分析 8
第三節 、情緒辭典 10
第四節 、斷詞服務 11
第三章 、研究流程 13
第一節 、建立資料集 14
第二節 、建立初始議題字典 15
第三節 、過濾資料 16
第四節 、擴增議題字典 16
第五節 、建立情緒字典 19
第六節 、識別文章之議題特徵與情緒 19
第四章 、彙總與觀察 24
第一節 、資料集說明 24
第二節 、議題字典設定與說明 25
第三節 、情緒分析結果與觀察 26
第五章 、結論與未來展望 36
參考文獻 37
附錄 40

壹、 中文部分
黃楷元. (2006). 電子布告欄中的政治意見呈現——以台大批踢踢實業坊政治類看板為例.
黃學碩. (1996). 網際網路公共傳播內容的分析--以交大 BBS 站為例.
廖達琪, 林福仁, 黃郁慈, 劉子昱, & 李承訓. (2012). 台灣立法委員政見資料庫之建置. 選舉研究, 19(2), 129-158.
劉從葦. (2006). 台灣政黨的政策位置: 非介入式與介入式測量的比較研究. 臺灣政治學刊, 10(2), 3-62.

貳、 西文部分
Agarwal, Apoorv, Xie, Boyi, Vovsha, Ilia, Rambow, Owen, & Passonneau, Rebecca. (2011). Sentiment analysis of twitter data. Paper presented at the Proceedings of the Workshop on Languages in Social Media.
Bartels, Larry M. (1986). Issue voting under uncertainty: An empirical test. American Journal of Political Science, 709-728.
Chen, Keh-Jiann, & Liu, Shing-Huan. (1992). Word identification for Mandarin Chinese sentences. Paper presented at the Proceedings of the 14th conference on Computational linguistics-Volume 1.
De Marneffe, Marie-Catherine, & Manning, Christopher D. (2008). Stanford typed dependencies manual. URL http://nlp. stanford. edu/software/dependencies manual. pdf.
Ding, Xiaowen, Liu, Bing, & Yu, Philip S. (2008). A holistic lexicon-based approach to opinion mining. Paper presented at the Proceedings of the international conference on Web search and web data mining.
Esuli, Andrea, & Sebastiani, Fabrizio. (2006). Sentiwordnet: A publicly available lexical resource for opinion mining. Paper presented at the Proceedings of Language Resources and Evaluation Conference. Vol. 6, pp. 417-422.
Jakob, Niklas, & Gurevych, Iryna. (2010). Extracting opinion targets in a single-and cross-domain setting with conditional random fields. Paper presented at the Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing.
Jiang, Long, Yu, Mo, Zhou, Ming, Liu, Xiaohua, & Zhao, Tiejun. (2011). Target-dependent Twitter Sentiment Classification. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies-Volume 1 (pp. 151-160). Association for Computational Linguistics.
Kim, Soo-Min, & Hovy, Eduard. (2006). Extracting opinions, opinion holders, and topics expressed in online news media text. Paper presented at the Proceedings of the Workshop on Sentiment and Subjectivity in Text.
Klein, Dan, & Manning, Christopher D. (2003). Accurate unlexicalized parsing. Paper presented at the Proceedings of the 41st Annual Meeting on Association for Computational Linguistics-Volume 1.
Ku, Lun‐Wei, & Chen, Hsin‐Hsi. (2007). Mining opinions from the Web: Beyond relevance retrieval. Journal of the American Society for Information Science and Technology, 58(12), 1838-1850.
Liu, Bing. (2012). Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, 5(1), 1-167.
Qiu, Guang, Liu, Bing, Bu, Jiajun, & Chen, Chun. (2011). Opinion word expansion and target extraction through double propagation. Computational linguistics, 37(1), 9-27.
Zhang, Ley, Ghosh, Riddhiman, Dekhil, Mohamed, Hsu, Meichun, & Liu, Bing. (2011). Combining lexiconbased and learning-based methods for twitter sentiment analysis. HP Laboratories, Technical Report HPL-2011, 89.