隨著Web 2.0的興起，網路社群漸漸成為人與人之間溝通的重要管道之一。在facebook塗鴉牆中，除了使用者自己可以主動留言外，還包含許多由遊戲或者是其他應用程式在使用者的塗鴉牆留言。在時間有限，但朋友許多的情況下，如何從一大群朋友之中，有效率地找出需要被關心的朋友，是件重要的事。在本論文中，系統首先過濾掉由其他程式所自動發布的留言，擷取使用者的朋友們近七日內的塗鴉牆留言，並將擷取出的留言與中文、英文、及表情符號的資料庫進行比對。其中，中文資料庫的建立是藉由眾包（crowdsourcing）的概念，我們在facebook上先製作了一個分詞遊戲，由登入該遊戲的使用者來協助進行中文字詞的情緒分類。所有蒐集的中文、英文、及表情符號在資料庫中都有對應的情緒類別。最後，系統將每篇留言的情緒量化為需要被關心的程度，每則留言與留言發布的時間結合，來給予每一位朋友需要被關懷的指標，再依照此關懷指標將朋友排序。此外，本系統也以視覺化曲線呈現每位朋友七日內的情緒波動。使用者可以直接點選分析過的節點，來直接連結對應的留言，與需要被關心的朋友互動。本系統已在facebook上開放連結使用，應用程式名稱為Friend_Smile。

With the population of Web 2.0, online community has gradually become one of the important channels when people communicating. The wall on Facebook not only includes users’ message but also but also includes many from the games or other apps, so that how to find friends who need desire to be caring in the short time which is very important. In this paper, system filter the messages that posted by other applications, just fetching friends of user who post messages by themselves during the past 7 days and also compare the messages with database including Chinese, English, and Emoticon. The method we construct the Chinese database is the concept of “crowdsourcing”. We build a Segment-Game application on Facebook, when users login this application, they can help. All of collecting database, such as Chinese, English, and Emoticon can be belonged to a catalog. Last part, each message can be qualified and after composing each message and post time, giving an Em_value for user to take care. This system gives an attention index of each friend, and according to this value to rank. Besides, this system shows friends’ emotion changes by diagram of curves during the past seven days. If users want to care for their friends, they can click the analyzed node to interacting with their friends. The system has been published on Facebook, and named “Friend_Smile”.

第一章 緒論.................................................1
1.1 研究背景和動機...........................................1
1.2 研究目的和重要性..........................................4
1.3 論文組織章節.............................................6
第二章 背景知識與相關研究......................................8
2.1 自然語言處理.............................................8
2.2 中文斷詞系統............................................10
2.3 眾包..................................................13
2.4 社群網路...............................................14
2.4.1 Facebook............................................14
2.5 社群情緒...............................................16
第三章 系統架構與方法........................................18
3.1 系統架構...............................................18
3.2 資料擷取...............................................19
3.2.1 前置處理............................................ 20
3.2.2 詞類標記處理......................................... 21
3.3 資料庫建立.............................................22
3.3.1 資料庫設計...........................................24
3.3.2 情緒資料建立..........................................26
3.4 中文情緒類別詞處理.......................................28
3.5 留言關懷類別值..........................................30
3.6 關懷指標排名............................................32
第四章 系統實作與評估........................................34
4.1 開發平台...............................................34
4.2 系統介面...............................................35
4.3 系統評估...............................................42
第五章 結論與未來研究方向.....................................48
參考文獻...................................................51
附錄 1 中研院平衡語料詞類標記集................................56
附錄 2 Facebook 權限.......................................58
附錄 3 英文論文.............................................64

圖目錄
圖1：Maslow need-hierarchy theory.........................17
圖2：系統架構...............................................19
圖3：系統的ERD 圖...........................................24
圖4：分詞遊戲...............................................27
圖5：XAMPP 控制版面.........................................34
圖6：Friends_Smile 登入頁面.................................36
圖7：Friends_Smile 授權請求.................................36
圖8：選擇關懷群組............................................37
圖9：朋友清單...............................................38
圖10：塗鴉牆留言判別範例......................................39
圖11：圖形化心情波動圖1......................................41
圖12：圖形化心情波動圖2......................................41

表目錄
表1：情緒類別和關懷類別值對照表................................29
表2：日期權重...............................................33
表3：Em_value 範例.........................................40
表4：英文詞彙評估............................................43
表5：表情符號評估............................................44
表6：中文評估...............................................45
表7：Em_value 評估.........................................46
表8：Attention_Index 評估..................................47

[1]	S. Milgram, “The Small World Problem,” Psychology Today, vol. 2, 1967 pp. 60-67,.
[2]	“Alexa the Web Information Company,” http://www.alexa.com/ (last accessed Jun 07, 2011).
[3]	K. J. Chen, & S.H. Liu, “Word Identification for Mandarin Chinese Sentences,” Proceedings of COLING, 1992, pp. 101-107.
[4]	K. J. Chen, & M. H. Bai, “Unknown Word Detection for Chinese by a Corpus-based Learning Method,” International Journal of Computational linguistics and Chinese Language Processing, vol.3, no.1, 1998, pp. 27-44.
[5]	K. J. Chen & W. Y. Ma, “Unknown Word Extraction for Chinese Documents,” COLING, 2002, pp. 169-175.
[6]	W. Y. Ma & K. J. Chen, “A bottom-up Merging Algorithm for Chinese Unknown Word Extraction,” ACL workshop on Chinese Language Processing, 2003, pp. 31-38. 
[7]	C. T. Li, H. C. Lai, C. T. Ho, C. L. Tseng and S. D. Lin, “Pusic : Musicalize Microblog Messages for summerization and Exploration,” 19th international conference on World wide web of ACM, April 26-30, pp. 1141-1142, 2010.
[8]	J. Bollen, H. Mao, and X. J. Zeng. “Twitter mood predicts the stock market,” Technical Report arXiv, CoRR, 2010.
[9]	N. H. Frijda, P. Ekman, R. J, and Davidson, “Varieties of Affect: Emotion and Episodes, Moods, and Sentiments,” The nature of emotion, New York, Oxford University Press, pp. 59-67, 1994.
[10]	“MSN Emoticons,” http://messenger.msn.com/Resource/Emoticons.aspx (last accessed Jun 07, 2011). 
[11]	“Complete List of Plurk Hidden Emoticons halfmadness,” http://otnaikzir.wordpress.com/2010/12/24/complete-list-of-plurk-emoticons (last accessed Jun 07, 2011).
[12]	“Emoticons - Yahoo! Messenger,” http://messenger.yahoo.com/Grammatical s/emoticons (last accessed Jun 07, 2011).
[13]	張春興, “現代心理學,” 東華書局, 3月, 1995年.
[14]	Z. Y. Sheu, “The Relationships among Strain Factors, Negative Emotion, Locus of Control, and Adolescent Device,” National Cheng Kung University, 2005.
[15]	R. B. Denhardt, J. V. Denhardt, and M. P. Aristigueta,  “Managing human behavior in Public and Non-profit organizations,” U.S.A: SAGE Publications, pp. 148-149
[16]	“Facebook Marketing Statistics, Demographics, Reports, and News - CheckFacebook,” http://www.checkFacebook.com (last accessed Jun 07, 2011).
[17]	Y. C. Lin, “Emotion Classification System based on Semantic Content Analysis,” Master thesis, National Cheng Kung University, 2003.
[18]	G. A. Miller, R. Beckwith, C. D. Fellbaum, D. Gross, and K. Miller, “WordNet: An online lexical database,” Int. J. Lexicograph, 1990, pp. 235-244.
[19]	C. Strapparava, A. Valitutti, “WordNet-Affect: an Affective Extension,” LREC’04, 2004, pp. 1083-1086.
[20]	V. Bobicev, V. Maxim, T. Prodan, N. Burciu and V. Angheluş, “Emotion in words: Developing a multilingual WordNet-Affect,” Computational Linguistics and Intelligent Text Processing Lecture Notes in Computer Science, vol. 6008, 2010, pp.375-384
[21]	C. Yang, K. H. Y. Lin, and H. H. Chen, “Building Emotion Lexicon from Weblog Corpora,” the 45th Annual Meeting of ACL, 2007, pp. 133-136.
[22]	C. Quan, F. Ren, “Construction of a Blog Emotion Corpus for Chinese Emotional Expression Analysis,” the 2009 Conference on Empirical Methods in Natural Language Processing, Singapore, 2009, pp. 1446-1454.
[23]	 “The Rise of Crowdsourcing,” http://www.wired.com/wired/archive/14.06/crowds.html (last accessed Jun 07, 2011).
[24]	D. C. Brabham, “Crowdsourcing as a Model for Problem Solving: An Introduction and Cases”, Convergence: The International Journal of Research into New Media Technologies, vol.14, no.1, 2008, pp. 75-90.
[25]	陳建州, “噗浪玩出大生意 = Plurk your brand plurk your business,” 時報出版, 10 月, 2009 年.
[26]	“Getting Started - Facebook Developers,” http://developers.Facebook.com/docs/ (last accessed Jun 07, 2011).
[27]	A. J. Kim, “Emergent Purpose”, Musings of a Social Architect, 2004.
[28]	A. H. Maslow “Motivation and Personality,” New York: Harper & Row, 1969.
[29]	 “apache friends - xampp,” http://www.apachefriends.org/ (last accessed Jun 07, 2011).