本論文中，提出不同於內容過濾與協同過濾之推薦系統，以社會導覽(Social Navigation)為基本概念，將群組內使用者瀏覽過的網頁進行推薦。對系統而言，利用初步的過濾以降低後端儲存以及計算的時間。 

在使用者訓練中，首先透過訓練，分析使用者瀏覽行為。利用訓練後使用者所獲得本身興趣的所屬類別，作為網頁投票的依據。在網頁的過濾上，透過使用者的瀏覽行為，將所蒐集的網頁進行所屬類別的投票，進而推薦。推薦上，以toolbar 作為與使用者的溝通方式。讓使用者在瀏覽器上安裝toolbar，就可以得到其他使用者的推薦資訊。

本論文的貢獻在於能改善Collaborative Filtering在處理效率上缺點，提供使用者對於自己感興趣的網頁類別中，其他使用者具有相同類別的網頁資訊。

In the following research, using the concept of social navigation, the web pages browsed by users are used to recommend web pages to each other.

 After the analysis of the history of recently viewed web pages, we can learn more about the user’s interests. In the filtering of web pages, we can set up a voting method in the target class mechanism, to recommend web pages to other users that have similar interests. The toolbar is the main medium of communication with the users. The user can immediately get recommended web pages on the browser.

 The main contribution of this paper is to improve collaborative filtering, and to provide web pages for other users with similar interests. As for the system, the initial filtering would lower the time required for calculation as well as the amount that needs to be saved.

目錄
第1章	緒論	1
1.1	研究動機與目的	2
1.2	研究內容	3
1.3	論文內容大綱	4
第2章	背景知識與相關研究	5
2.1	資訊檢索(Information Retrieval)	6
2.1.1	二元模式(Boolean Model)	6
2.1.2	向量空間模式(Vector Space Model)	7
2.1.3	機率模式(Probability Model)	9
2.2	推薦系統	11
2.2.1	內容過濾(Content-Based Filtering)	12
2.2.2	協同過濾(Collaborative Filtering)	13
2.3	代理伺服器(Proxy Server)	14
2.4	自動分類演算法	17
2.4.1	最近鄰居決定法則 (K-Nearest Neighbor rule)	17
2.4.2	貝式分類器 (Bayes Classifier)	19
2.4.3	特徵鑑別公制(Discriminative Feature Metric)	20
2.5	英國國家標準語料庫	23
第3章	系統架構	26
3.1	使用者瀏覽行為訓練	28
3.1.1 網頁分析	28
3.1.2 使用者分類	30
3.2	網頁投票	32
3.3	推薦機制	34
第4章	實驗步驟與結果	36
4.1	實驗步驟	36
4.2	實作介面	37
4.3	推薦效果評估	38
第5章	結論與未來研究	39
5.1	結論	39
5.2	未來研究方向	40
參考文獻	41
附錄	44
 
圖目錄
圖2.1 The cosine of Ө is adopted as  	9
圖2.2 Proxy Server 的運作原理簡介	14
圖2.3投票機制	22
圖3.1 系統架構圖	26
圖3.2 前處理簡易流程圖	28
圖3.3 關係網頁資料庫	30
圖3.4 使用者分類圖	31
圖3.5 網頁投票機制圖	32
圖3.6 推薦機制	35
圖3.7 資料表參照圖	35
圖4.1 Open Directory Project網站	36
圖4.2 使用介面	37
 
表格目錄
表2.1 BNC語料內容分類表	25
表2.2 BNC語料與年齡關係表	25
表4-1 投票結果	38

[1] R. Baeza-Yates and B. Ribeiro-Neto, “Modern Information Retrieval”, Addison Wesley Longman, Inc, May 1999.
[2] G. Salton and M. J. McGill, “Introduction to Modern Information Retrieval”, McGraw-Hill Book Co., New York, 1983.
[3] S. E. Robertson and K. Sparck Jones, “Relevance Weighting of search terms”,  Journal of the American Society for Information Sciences, 27(3), 1976, pp.129-146.
[4] A. Ansari, S. Essengaier, and R. Kohli, “Internet Recommendation Systems”, Journal of Marketing Research, 37(3), pp.363-375, Aug. 2000.
[5] Schafer, J.B., Konstan, J.A., and Riedl, J., “Electronic Commerce Recommender Applications”, Journal of Data Mining and Knowledge Discovery, 5(1/2), pp.115-152. 2001.
[6] M. Pazani, J. Muramatsu, and D. Billsus. “Syskill & Webert: Identifying interesting Web sites”, In Proc. of the National Conference on Artificial Interlligence, AIII, pp.54-61, 1996.
[7] J. Rucker and M.J. Polanco, “Siteseer: personalized navigation for the Web”, Communications of the ACM, 35(12), pp.73-75, Dec.1992.
[8] B. Krulwich and C. Burkey, “Learning user information interests through extraction of semantically significant phrases”, In Proc. of the AAAI Spring Symposium on Machine Learning in Information Access, Stanford, CA, 1996.
[9] Lang, K. “NewsWeeder: Learning to filter Netnews”, In Proc. of the Twelfth International Conference on Machine Learning, pp.331-339 1995.
[10] M.J. Pazzani, “A framework for collaborative, content-based and demographic filtering”, Artificial Intelligence Review, pp.393-408, 1999.
[11] R.J. Mooney and L. Roy, “Content-based book recommending using learning for text categorization”, In Proc, of the 5’th ACM Conf. on Digital Libraries, pp.195-240, June 2000.
[12] J.S. Breese, D. Heckerman and C. Kadie, “Empirical analysis of predictive algorithms for collaborative filtering”, Tech. Report, MSR-TR-98-12, Microsoft Research, Oct. 1998.
[13] Yu, C., Ooi, B.C., Tan, K.L., Jagadish, H.V, “Indexing the Distance: An Efficient Method to KNN Processing”, VLDB, 2001.
[14] Christopher D. Manming, Hinrich Schutze(2000), “Foundations of Statistical Natural Language Processing”, 3rd ed,pp232-249
[15] X. Liu and Y. Gong, “Document clustering with cluster refinement and model selection capabilities”, In Proceedings of ACM SIGIR 2002, Tampere, Finland, Aug. 2002.
[16] BNC - British National Corpus.
[17] DOM: Document Object Model  http://www.w3.org/DOM/
[18] Suhit Gupta & Gail Kaiser & David Neistadt & Peter Grimm “DOM-based Content Extraction of HTML”, Documents WWW 2003
[19] Lan Yi & Bing Liu & Xiaoli Li “Eliminating Noisy Information in Web Pages for Data Mining” , SIGKDD 2003.
[20] Salton G. and Buckley G., “Term-weighting approaches in automatic text retrieval”, Information Processing and Management, 24, pp513-523, 1988
[21] Dieberger, Andreas “Supporting Social Navigation on the World Wide Web”, smartech.gatech.edu -1997.
[22] A. Dieberger, P. Dourish, K. Höök, P. Resnick, A. Wexelblat, “Social navigation: techniques for building more usable systems”, interactions 2000
[23] Open Directory Project (ODP). http://dmoz.org
[24] Ming-Feng Lu , “The Design of an Agent for Finding Related Web Pages and its Application to English Learning”, Department of Computer Science and Information Engineering of Tamkang University, June 2006
[25] Chia-Chun Peng, “The Designing of a Web Page Recommendation System for ESL”, Department of Computer Science and Information Engineering of Tamkang University, January, 2007