近年來隨著Web2.0概念的發展，使得部落格、微網誌、社群網站等得以快速興起且廣為接受與喜愛，尤其是微網誌提供了簡單易用及快速回應等特性，漸漸改變了使用網路服務的習慣及人際互動的方式。由於微網誌提供了獨特的朋友關連環境，使用者抒發的情緒及發表的近況很容易與朋友之間分享與傳播，成為鼓勵打氣、尋求心靈安慰及互動的重要管道。基於以上原因，本論文從微網誌擷取使用者之文字訊息，並希望有別於現在從微網誌語料中對情緒類別做辨識之相關研究，利用機器學習的方法針對使用者隱藏在文字內產生情緒困擾的原因來做辨識，並且使用新竹教育大學圖書館心靈驛站專區之典藏，配合諮商專家結合書目療法之理論，針對使用者藏著情緒困擾的留言，用噗浪書目治療推薦系統回應適合使用者情緒之書籍。
最後，實驗期間系統蒐集了3823篇的留言，系統在當中辨識出100篇是帶有情緒原因的留言來回覆訊息及推薦書目，也從後續觀察中得知使用者對於系統回覆訊息的感覺以及系統分類的正確率，此外利用問卷評估了使用者對於推薦的書籍的喜好度、推薦系統的介面設計度、使用者成效等來做為系統的成效判定，每個項目分別至少都有達70%以上的滿意度，有效的達成當初設定的目的。

With the concept of Web2.0 developing in recent years, blogs, micro-blogs, and community site rise rapidly and are accepted. Especially, the features of the micro-blogs such as easily used and fast replied, have changed people whose habits of network service and interaction way little by little. Because micro-blogs provide the unique environmentof the friend connections, it can be used to share and dispread the emotions and the recent status which users post between friends.It becomes an important tract to look for comfort,be encouraged, and interaction with others.
This thesis proposed a methodology for recommending books which are related with readers’ emotions on-line. With the assistance of machine learning technologies, readers’ emotions could be identified automatically by analyzing the text on readers’ micro-blogs.The book collections of“TherapyArea” in National Hsin-Chu University of Education Library support hundreds of books for expert when consulting with readers.
3823 sheets of messageswas collected between the experiment, 100 sheets related with emotion reasons were recognized correctly and repliedautomatically with the messages of books recommending. Readers’ satisfactions of books recommendation were evaluatedby questionnaires. Some feedbacks of the books recommending result, user interface and the users’satisfactionswere also discussed at the end of thesis. At least of 70% of readers reflected that the system recommended books which were adaptive with their emotions on-time.

中文參考文獻
1. 董氏基金會, 大學生主觀壓力源與憂鬱情緒之相關性調查.大家健康, 2008(10): p. 54-58.
2. 陳書梅, 書目療法、公共圖書館與青少年之療癒閱讀。.臺灣圖書館管理季刊, 2010. 6(1).
3. 洪新柔, 大學圖書館發展性書目療法服務需求之研究 ─以國立臺灣大學大學部學生為例, in 圖書資訊學研究所. 2009, 臺灣大學: 臺北市. p. 230.
4. 林豪鏘, 電子商務與企業電子化 -- 從 Web2.0 到 U 化商務. 第三版 ed. 2010: 旗標.
5. 林宇中, 基於語意內容分析之情緒分類系統, in 資訊工程學系碩士班. 2003, 國立成功大學.
6. 周嵩能, 以微網誌語料進行情緒辨識之研究, in 資訊工程學系. 2010, 國立台南大學.
7. 王偉哲, 結合交互訊息與語意線索之情緒辨識機制, in 數位學習科技學系碩士班. 2010, 國立台南大學.
8. 陳書梅, 圖書館與書目療法服務.大學圖書館, 2008. 12(2): p. 15-41.
9. 陳書梅, 後 SARS 時代與書目療法. 臺北市立圖書館館訊, 2003. 4(20): p. 78-82.
10. 田宜歆, 輔助教師以機器學習分析及評量學生鋼琴演奏技巧之研究 2008.

英文參考文獻
11. Forrest, M.E.S., Recent developments in reading therapy: A review of the literature. Health Libraries Review, 1998(15): p. 157-164.
12. Usherwood, B., &Toyne,J., The value and impact of reading imaginative literature. Journal of Librarianship and Information Science, 2002. 34(1): p. 33-41.
13. Hynes, A.M.H.-B., M., biography:the interactive process. 1986: Boulder,CO:Westview.
14. Cohen, L.J., The therapeutic use of reading: A qualitative study. Journal of Poetry Therapy, 1993(7): p. 73-83.
15. Bibliotherpy, in Webster's third new international dictionary of the English language, unabridged and britannica world language dictionary, Merriam, Editor. 1961, Springfield: Mass.
16. Rubin, R.J., Using bibliotherapy: A guide to theory and practice. 1978, London: Oryx.
17. Baruchson-Arbib, S., Social information science: Love health and the information society. 1996, Brighton: Sussex Academic.
18. Alexa. Available from: http://www.alexa.com/.
19. Cornett, C.E.C., C. F., Bibliotherapy: The right book at the right time. 1980, Indiana, Bloomington: Phi Delta Kappa Educational Foundation.
20. Pardeck, J.A., & Pardeck, J. T., Young people with problems: A guide to bibliotherapy. 1984, Westport,London: Greenwood.
21. Riordan, R.J., & Wilson, L. S., Bibliotherapy: Does it work? Journal of Counseling and Development, 1989. 9(67): p. 506-508.
22. Plurk API. Available from: http://www.plurk.com/API.
23. N.Thao, I.B., L. Mingkun, and I. K. Sethi,” Investigation of combining SVM and decision tree for emotion classification, “in Multimedia, Seventh IEEE International Symposium on, 2005, p. 5.pp.
24. Quinla, J.R., C4.5: programs for machine learning. 1993: Morgan Kaufmann Publishers Inc. . 302.
25. Feigenbaum, R.A. The art of artificial intelligence: I. Themes and case studies of knowledge engineering. in Proc. of the 5th IJCAI. 1977.
26. Lun-Wei Ku , H.-H.C., Mining Opinions from the Web: Beyond Relevance Retrieval. Journal of American Society for Information Science and Technology, 2007(Special Issue on Mining Web Resources for Enhancing Information Retrieval): p. 1838-1850.