臺灣博碩士論文加值系統

由於推薦系統近年來在各項領域中受到廣泛的使用，在各大商務網站中也受到相當程度的重視，而推薦系統中最常被運用的協同過濾式推薦，更是廣泛地被運用在電子商務領域中。傳統的推薦系統都是針對商品的量化回饋或交易資訊來進行資料分析，再依此數據分析進行內容導向式(Content-Based)推薦或協同過濾式(Collaborative Filtering)推薦。不過當面臨產品項目過多或者使用者數量多的情形下，要找到和目標使用者一樣曾購買過類似產品的其它使用者，很有可能相關資料並不會太多，因此會面臨可參考的群族成員較少，致使推薦較變得更加困難。
現今社群網路(Social Network)的盛行，許多使用者透過社群平台對產品進行評價及評論，社群成員也會在彼此間透過分享、留言、按讚、回饋等行為，來加深彼此間的交流和關係，經由這樣的互動進而產生網路社群成員之間的信任關係。許多學者們也將這樣的關係透過一些計算的方法來找出其信任值。
有關使用者對產品的留言回饋意見，過去的研究通常較側重於語意分析(semantic analysis)或字詞解析等議題，較少有研究將使用者評論內容之正向/負向意義加以運用分析。事實上，使用者的回饋意見通常含有對產品正向或負向的評價，這類評價更能代表使用者或相關的留言者他們對產品的共通看法，因此有相同留言看法的使用者們，其偏好的相似性，也具有推薦的參考價值。為了找到更多的可能被推薦的潛在使用者，本研究認為除了傳統的推薦方法外，亦應加入使用者對產品的評論或留言加以分析，找出使用者會與其他使用者對產品留言之間的信任關係。有鑑於此，本研究提出一個以網路評論做為信任基礎的推薦系統，本研究擷取使用者對產品之評論或對其它使用者的評論留言等，做為信任分析依據，藉此來擴增可進行推薦的使用者群組。
本研究以amazon.com網站使用者對產品的評論和回應意見，先建立出一個以信任為模型的網路，接著透過TF-IDF將這些評論進行字詞解析及向量化，並利用SVM對這些評論的內容進行分類(正向/負責)，找出使用者之間的信任關係，據此作為擴充推薦系統可參考的偏好對象的方法。藉此方法擴增推薦的廣度並提升資料覆蓋率，找出更多的潛在用戶，並將更多用戶可能有興趣的產品推薦給目標用戶。

Recommendation system have been widely used in various fields in recent years and have received considerable attention in major business website. The most frequently used collaborative filtering recommendation is extensively applied to electronic commerce. While conventional product recommendation system tend to analyze the quantitative feedbacks or information of transactions, and the analysis of feedbacks focuses more on the semantics, the message and the correlation between them, the reply of feedbacks are less discussed.
In view of this, we would like to present an analysis of correlation among users’ product feedbacks. By the possible trust among the users who comment on the product, this research try to figure out if it is possible to make such comments for recommendation.
In this study, we set up a network of a trust-based model according to comments from the responses of Amazon (amazon.com), quantify them by using TF-IDF Comments, use SVM to classify this quantitative information, and identify the relation of trust among users. Then, we further adjust the users’ trust node with In-Degree weight(s) and increase confidence values as the basis for recommendation. Making it a method of collaborative filtering recommendation, this research thereby adds the recommendation accordance, improves data coverage, find potential users, and recommend the products that a target user may be interested in.

摘要………............................................................................................ I
Abstract………....................................................................................... III
誌謝………............................................................................................ IV
第壹章、導論………............................................................................ 1
第貳章、文獻探討……….................................................................... 4
2.1 推薦系統………........................................................................ 4
2.2 協同過濾式推薦系統……….................................................. 7
2.3 信任值 (Trust Value)………................................................... 10
2.4 初始信任值修正 (Initial Trust Value)..................................... 12
2.5 信任值修正(Trust Updating).................................................. 14
2.6 信任推演(Trust Propagation).................................................. 15
2.7 支持向量機 (Support Vector Machine, SVM)........................ 19
2.8 字頻分析TF-IDF (term frequency–inverse document
frequency)……………………………..........................……… 21
第參章、研究方法............................................................................... 24
3.1 研究流程................................................................................. 25
3.2 評論資料擷取......................................................................... 26
3.3 評論字詞分析......................................................................... 26
3.4 TF-IDF評論資料特徵擷取.................................................... 27
3.5 使用SVM進行評論分類....................................................... 28
3.6 信任值計算…………………………………………………. 30
3.7 信任推演……………………………………………….......... 31
第肆章、實驗過程與結論................................................................... 36
4.1 資料蒐集(Data set)............................................................. 36
4.2 計算信任初始值................................................................. 38
4.2.1 TF-IDF擷取評論特徵向量............................................... 38
4.2.2 SVM分類準確度測量....................................................... 39
4.3 建立使用者信任網路......................................................... 40
4.3.1 In-Degree權重值計算........................................................ 41
4.3.2 In-Degree權重值結合推演方法........................................ 42
4.4 實驗設計………………………......................................... 42
4.5 實驗結果…………………………………………….. 43
第伍章、結論與未來研究.................................................................... 50
參考文獻................................................................................................. 52

圖目錄
圖 1：信任網路 12
圖 2：正向信任網路示意圖 13
圖 3：直接推演 16
圖 4：共同信任 16
圖 5：信任轉換 17
圖 6：耦合信任 17
圖 7：信任傳遞 18
圖 8：支持向量 20
圖 9：本研究進行之步驟與流程 25
圖 10：In-Degree 權重計算示意圖 31
圖 11: 信任值推演各項公式之數值計算分佈範例 33
圖 12：信任強度高使用者(E)的關聯值增強示意圖 34
圖 13：信任值的增強若以 Sk=0.5, 0.75, 0.9 為例，則對應的 Tu,k值變化圖………………………………………………..…... 35
圖 14：使用者對產品的評分與評論……………………………… 37
圖 15：其他使用者討論與評論…………………………………… 37
圖 16：特徵向量計算流程………………………………………… 38
圖 17：信任網路關聯圖(僅摘要其中部份節點)………….……… 40
圖 18：第一組實驗結果圖表...……………………………............. 44
圖 19：第二組實驗結果圖表...……………………………............. 45
圖 20：第三組實驗結果圖表...……………………………............. 46
圖 21：第四組實驗結果圖表............…………………………….... 47
圖 22：第五組實驗結果圖表............................................................ 48

表目錄
表 1：推薦技術分類表 6
表 2：協同過濾式推薦相關應用……….......................................... 9
表 3：SVM 訓練資料之範例說明…………………………………. 28
表 4：回應率與準確率…………………………………………….. 39
表 5：In-Degree 權重值計算結果(僅摘要其中部份節點)……….. 41
表 6：第一組實驗結果...................................................................... 44
表 7：第二組實驗結果...................................................................... 45
表 8：第三組實驗結果...................................................................... 46
表 9：第四組實驗結果...................................................................... 47
表 10：第五組實驗結果 48
表 11：覆蓋率比較 49
表 12：信任關係人數比較 49

[1] 陳寶旭，「運用協同過濾推薦機制於搭售商品之推薦」，朝陽科技大學資訊管理系碩士論文，2006年。
[2] 廖品妍，「以顯性評價為主之相似性推薦」，朝陽科技大學資訊管理系碩士論文，2010年。
[3] 蔡美慧，「以偏好導向為相似性量測的協同過濾式推薦」，朝陽科技大學資訊管理系碩士論文，2010年。
[4] 許榮望，「評論語意自動化分析與顧客滿意度應用之研究-以中國旅遊網站為例」，朝陽科技大學資訊管理系博士論文，2013年。
[5] 遊和正、黃挺豪、陳信希，「領域相關詞彙極性分析及文件情緒分類之研究」，中文計算語言學期刊，第17 卷第四期，第33-48頁，2012年。
[6] 侯彥旭，「以平價理論為基礎的電子本文分類」，中原大學資訊管理學系碩士論文，2010年。
[7] 鍾任明、李維平、吳澤民，「運用文字探勘於日內股價漲跌趨勢預測之研究”，中華管理評論，第十卷第一期，第1-30頁，2007年。
[8] 黃河銓、李奕縉、黃昱凱，「整合信任網路與回饋機制之個人化餐廳推薦系統」，資訊與管理科學，5(2)，第65-78頁，2012年。
[9] 馬傳淵，「於社群網路上結合共同評價及信任關聯以進行協同推薦」，國立中山大學資訊管理學系研究所碩士論文，2014年。
[10] 楊昇翰，「以社群網評論分析作為信任基礎之推薦系統」，朝陽科技大學資訊管理系碩士論文，2016。
[11] Abdul-Rahman Alfarez and Stephen Hailes, “Supporting Trust in Virtual Communities,” Proceedings of the 33rd Annual Hawaii International Conference on System Sciences, Hawaii, pp. 1-9, 2000.
[12] Gediminas Adomavicius and Alexander Tuzhilin, “Towards the Next Generation of Recommender Systems--A Survey of the State-of-the-Art and Possible Extensions,” IEEE Transactions on Knowledge and Data Engineering, vol. 17, no. 6, pp. 734-749, 2005.
[13] Bernhard E. Boser, Isabelle M. Guyon, and Vladimir N. Vapnik, “A Training Algorithm for Optimal Margin Classifiers,” The 5th Annual ACM Workshop on COLT, New York, pp. 144-152, 1992.
[14] J. Breese, D. Heckerman, and C. Kadie, “Empirical analysis of predictive algorithms for collaborative filtering,” Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence, pp. 43-52, 1998.
[15] Federico Chesani, “Recommendation Systems,” Corso di laurea in Ingegneria Informatica, 2002.
[16] Nello Christianini and John Shawe-Taylor, “An Introduction to Support Vector Machines and Other Kernel-based Learning Methods,” Cambridge, U.K.: Cambridge Univ. Press, vol. 18, no. 6, pp 687-689, 2000.
[17] Susan Dumais, John Platt, and David Heckerman, “Inductive Learning Algorithms and Representations for Text Categorization,” Proceedings of the 7th International Conference on Information and Knowledge Management, pp. 148-155, 1998.
[18] Abhinandan Das, Mayur Datar and Ashutosh Garg, “Google News Personalization: Scalable Online Collaborative Filtering,” Proceedings of 16th International Conference on World Wide Web, Canada, pp. 271-280, 2007.
[19] David Goldberg, David Nichols, Brian M. Oki, and Douglas Terry, “Using collaborative iltering to weave an information tapestry,” Communications of the ACM, Volume 35, Issue 12, pp61-70, 1992.
[20] Jennifer Golbeck, “Generating Predictive Movie Recommendations from Trust in Social Network,” Proceedings of The 4th International Conference on Trust Management Lecture Notes in Computer Science, vol. 3986, pp. 93-104, 2006.
[21] J. Golbeck. “Generating predictive movie recommendations from trust in social networks. ” Proceedings of the fourth International Conference on Trust Management, 2006.
[22] J. Golbeck. “Computing with trust: Definition, properties, and algorithms. ” Securecomm and Workshops, 2006.
[23] R. Guha , Ravi Kumar, “Propagation of Trust and Distrust,” Proceedings of the 13th International Conference on World Wild Web, New York, pp. 403-412, 2004.
[24] Chein-Shung Hwang and Yu-Pin Chen, “Using Trust in Collaborative Filtering Recommendation,” New Trends in Applied Artificial Intelligence, Vol. 4570, pp. 1052-1060, 2007.
[25] Sheng-Han Yang, Incheon Paik, Li-Hua Li, “Improving Trust Based Recommender System By Mining User Comments on the Web,” The 2015 World Congress on Information Technology Applications and Services, Jeju, Korea, 2015.
[26] Xuejiao Han, Wenqian Shang, Shuchao Feng, (2015), “The Design and Implementation of Personalized News Recommendation System” IEEE/ACIS 14th International Conference on Computer and Information Science (ICIS), Las Vegas.
[27] Santosh.D Teja and Babu. K Sudheer, “Building Trust In Online Reviews by a Customer through Linked Data.” IEEE Region 10 Symposium, pp. 342-346, 2014.
[28] U. Shardanand and P. Maes, “Social information filtering: algorithms for automating “word of mouth”,” Proceedings of the SIGCHI Conference on Human Factors inComputingSystems, pp. 210-217, 1995.
[29] Mohsen Jamali and Martin Ester, “A Matrix Factorization Technique with Trust Propagation for Recommendation in Social Networks,” Proceeding of the 4th ACM Conference on Recommender Systems, Barcelona, pp. 135-142, 2010.
[30] Banage T. G. S. Kumara, Incheon Paik, Hiroki Ohashi, Yuichi Yaguchi and Wuhui Chen, “Context-Aware Filtering and Visualization of Web Service Clusters,” Proceedings of IEEE International Conference on Web Services (ICWS), Alaska, pp. 89-96, 2014.
[31] B. Kim, Q. Li, C. Park, S. Kim, and J. Kim, “A new approach for combining content-based and collaborative filters,” Journal of Intelligent Information Systems, Vol. 27, No. 1, pp. 79-91, 2006.
[32] K. Kwon, J. Cho, and Y. Park, “Multidimensional credibility model for neighbor selection in collaborative recommendation,” Expert Systems with Applications, Vol. 36, No. 3, Part 2, pp. 7114-7122, 2009.
[33] Neal Lathia, Stephen Hailes and Licia Capra, “Trust-Based Collaborative Filtering,” Proceedings of IFIPTM 2008: Joint iTrust and PST Conferences on Privacy, Trust Management and Security, June 18-20, 2008, Trondheim, Norway, pp. 119, 2008.
[34] Nan Li, Desheng Dash Wu, “Using text mining and sentiment analysis for online forums hotspot detection and forecast,“ Decision Support Systems, vol. 48, no. 2, pp. 354–368, 2010.
[35] Wei-Po Lee and Chuan-Yuan Ma, “Enhancing collaborative recommendation performance by combining user preference and trust-distrust propagation in social networks,” Knowledge-Based Systems.
[36] Shan Liu, Yao Dong, Jianping Chai, (2016), “Research of Personalized News Recommendation System Based on Hybrid,” 2nd IEEE International Conference on Computer and Communications (ICCC), pp. 865-869.
[37] Paolo Massa and Paolo Avesani, “Trust-aware Recommender Systems,” Proceeding of the ACM Conference on Recommender Systems, Minnesota, USA, pp. 17-24. 2007.
[38] Samaneh Moghaddam, Mohsen Jamali, and Martin Ester, “Feedback Trust: Using Feedback Effects in Trust-based Recommendation Systems,” Proceedings of the 3th ACM Conference on Recommender Systems, New York, USA, pp. 269-272, 2009.
[39] S. Al-Oufi, H. N. Kim, and A. El Saddik. “A group trust metric for identifying people of trust in online social networks.” Expert Systems with Applications, 39(18): 13173-13181, 2012.
[40] John O’Donovan, Barry Smyth, “Trust in Recommender Systems,” IUI ’05 Proceedings of the 10th International Conference on Intelligent User Interfaces, New San Diego, pp. 167-174, 2005.
[41] Incheon Paik and Hiroshi Mizugai, “Recommendation System Using Weighted TF-IDF and Naïve Bayes Classifiers on RSS Contents,” Journal of Advanced Computational Intelligence and Intelligent Informatics, vol. 14, No. 6, 2010.
[42] John C. Platt, “Probabilistic Outputs for Support Vector Machines and Comparisons to Regularized Likelihood Methods,” Advances in Large Margin Classifiers, 1999.
[43] Michael J. Pazzani and Daniel Billsus, 2007, Content-Based Recommendation Systems, Springer-Verlag Berlin Heidelberg, LNCS 4321, pp. 325 – 341, 2007.
[44] P. Resnick, N. Iacovou, Mitesh Suchak, Peter Bergstrom, and John Riedl, “GroupLens: an open architecture for collaborative filtering of netnews,” Proceedings of the 1994 ACM Conference on Computer Supported Cooperative Work, pp. 175-186, 1994.
[45] Catarina Silva, Bernardete Ribeiro, “On text-based mining with active learning and background knowledge using SVM,” Soft Computing, vol. 11, no. 6, pp. 519-530, 2006.
[46] Georges Siolas, d’Alche-Buc Florence, “Support Vector Machine based on a Semantic Kernel for Text Categorization.” Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Network, Como, vol. 5, pp. 205-209, 2000.
[47] B. Sarwar, G. Karypis, J. Konstan, and J. Riedl, “Analysis of recommendation algorithms for e-commerce, ” Proceedings of the 2nd ACM Conference on Electronic Commerce, pp. 158-167, 2000.
[48] M. Sunitha, T. Adilakshmi, (2016), “Mobile Based Music Recommendation System,” International Conference on Inventive Computation Technologies (ICICT).
[49] Pree Thiengburanathum, Shuang Cang, Hongnian Yu, (2016), “Overview of Personalized Travel Recommendation Systems, ” 22nd International Conference on Automation and Computing (ICAC), Colchester .
[50] Vladimir Vapnik and Corinna Cortes, “Support vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–293, 1995.
[51] Jianshu Weng, Chunyan Miao and Angela Goh, “Improving Collaborative Filtering with Trust-based Metrics,” Proceedings of the 2006 ACM symposium on Applied computing (SAC '06), pp. 1860-1864, 2006.
[52] http://epinion.com (2014)
[53] http://amazon.com (2014)
[54] http://svmlight.joachims.org/