近年來在web-based社群網路上的推薦系統當中，加入使用者之間的信任關聯以產生更好的推薦績效，已經是一種非常普遍的方法。然而隨著各種線上社群網路的興起，使用者之間的關聯也不再侷限於單純的友好或正向信任關聯。許多的線上社群網站都在使用者的關聯上加入了負向信任的機制，如：Epinions.com可設定其他使用者為白名單(Trust)或黑名單(Distrust)、Slashdot.org可將其他使用者設定為朋友或敵人等等。因此，使用者之間的負向關聯對於推薦系統的影響具有其重要性。但在過去對於利用信任的傳遞性找出使用者之間隱含關聯的研究上，主要都著重於探討正向信任關聯的傳遞。而在負向信任的傳遞研究上，普遍認為負向信任的傳遞不如正向信任關聯單純，例如：我們能夠理解對於朋友的朋友也具有友好關聯，但卻難以辨認敵人的敵人是敵是友。因此認為負向信任無法像正向信任關聯般傳遞。
本研究根據社會心理學的結構平衡理論當中的穩定結構，提出符合負向信任傳遞的條件與方法，並設定多種實驗情境，各別測試正負向信任對於推薦系統績效的影響。除了使用者之間相對性的區域信任關聯外，本研究亦加入了屬於全域性信任度的In-Degree值進行實驗與比較，並利用信任值與In-Degree權重對於K個最近鄰居法的相似度進行調整，更利用K個最近鄰居法的TOP K鄰居，將In-Degree權重與信任值應用到矩陣分解演算法當中，有效針對矩陣分解演算法中的預測評分公式進行合理的修正。
根據實驗結果，可以看出負向信任在使用本研究提出的方法上，能夠成功傳遞其負向信任值，且對於推薦系統績效亦有幫助。而In-Degree值的影響也在考量負向In-Degree值的結合下有較佳的表現。

It becomes a more and more popular approach to develop recommendation systems based on trust relationships on the social networks. While social information that explicitly indicate the trust relationships among users has been used, some recent studies have started to use the transitivity of trust relationships to find the hidden relationships. However, the latter trend mainly makes use of the positive trust relationship; it considers distrust relationships as not transitive and thus not applicable to the recommendation systems.
Based on the structural balance theory of social psychology, in this thesis, I propose a new method that exploits the transitivity of distrust relationships with the positive trust relationships, and analyze the conditions for using these relationships in making recommendations. The proposed method integrates both of the trust and distrust values derived from the social networks and the user-similarities measured from the co-ratings of different users on items. With the hybrid measurements, the collaborative filtering techniques, including the k-nearest neighbor algorithm and the matrix factorization algorithm are used to predict the user preferences of un-rated items. A series of experiments have been conducted and the results show that the distrust relationships and the transitive effect among users can improve the performance of recommendation.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖次 vii
表次 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 3
第二章 文獻探討 5
2.1 全域信任與區域信任 5
2.2 信任關聯的概念 5
2.2.1 信任值傳遞計算 6
2.2.2 負向信任關聯 7
2.3 In-Degree研究 9
2.4 結合信任的協同推薦 9
2.4.1 結合全域信任的協同推薦 9
2.4.2 結合區域信任的協同推薦 10
2.5 協同過濾推薦演算法 10
2.5.1 K個最近鄰居法(k-nearest neighbor, KNN) 11
2.5.2 矩陣分解演算法(Matrix Factorization, MF) 12
第三章 研究方法 13
3.1 研究流程 13
3.2 In-Degree權重計算 14
3.2.1 PTRN的In-Degree權重 14
3.2.2 NTRN的In-Degree權重 16
3.3 信任關聯傳遞計算 19
3.3.1 正向信任關聯傳遞值計算 19
3.3.2 負向信任關聯傳遞值計算 21
3.4 資料集選擇 27
3.4.1 Flixster資料集 28
3.4.2 Epinions資料集 29
3.5 評估方法 30
第四章 實驗與結果 31
4.1 實驗設計 31
4.1.1 實驗情境介紹 32
4.1.2 資料前處理 36
4.2 實驗結果 37
4.2.1 KNN演算法 37
4.2.2 MF結合KNN演算法 44
4.3 結果分析 49
4.3.1 資料前處理分析 49
4.3.2 In-Degree權重分析 53
4.3.3 信任傳遞值分析 53
4.3.4 結合In-Degree與信任傳遞分析 54
4.4 結論 54
第五章 結論與未來研究 56
5.1 結論 56
5.2 未來研究 57
參考文獻 58

[1] M. Kalz, H. Drachsler, J. Van Bruggen, H. Hummel, and R. Koper, "Wayfinding services for open educational practices." International Journal of Emerging Technologies in Learning, 3(2): 1-6, 2008.
[2] N. Lathia, S. Hailes, and L. Capra. "The role of trust in collaborative filtering." http://www0.cs.ucl.ac.uk/staff/l.capra/publications/lathia_recsys_handbook09.pdf, 2009. online; accessed at Sep 3, 2013.
[3] J. Golbeck, and J. Hendler. "Inferring binary trust relationships in web-based social networks." ACM Transactions on Internet Technology (TOIT), 6(4): 497-529, 2006.
[4] J. Leskovec, D. Huttenlocher, and J. Kleinberg. "Predicting positive and negative links in online social networks." In Proceedings of the 19th International Conference on World Wide Web, 2010.
[5] R. Guha, R. Kumar, P. Raghavan, and A. Tomkins. "Propagation of trust and distrust." In Proceedings of the 13th International Conference on World Wide Web, 2004.
[6] J. Leskovec, D. Huttenlocher, and J. Kleinberg. "Signed networks in social media." In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2010.
[7] J. Golbeck. "Computing with trust: Definition, properties, and algorithms." In Securecomm and Workshops, 2006.
[8] J. Golbeck, B. Parsia, and J. Hendler. "Trust networks on the semantic web." In Proceedings of the 7th International Workshop on Cooperative Intelligent Agents, 2003.
[9] 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.
[10] D. Goldberg, D. Nichols, B. M. Oki, and D. Terry. "Using collaborative filtering to weave an information tapestry." Communications of the ACM, 35(12): 61-70, 1992.
[11] B. Mobasher, X. Jin, and Y. Zhou. "Semantically enhanced collaborative filtering on the web." In Web Mining: From Web to Semantic Web. Springer Berlin Heidelberg, 2004.
[12] X. Su, and T. M. Khoshgoftaar. "A survey of collaborative filtering techniques." Advances in Artificial Intelligence, 2009.
[13] L. P. Kaelbling, M. L. Littman, and A. W. Moore. "Reinforcement learning: A survey." arXiv preprint cs/9605103, 1996.
[14] P. Victor, C. Cornelis, M. De Cock, and A. Teredesai. "Trust-and distrust-based recommendations for controversial reviews." In Web Science Conference (WebSci'09: Society On-Line). No. 161, 2009.
[15] P. Victor, N. Verbiest, C. Cornelis, and M. De Cock. "Enhancing the trust-based recommendation process with explicit distrust." ACM Transactions on the Web (TWEB), 7(2), 2013.
[16] H. Ma, T. C. Zhou, M. R. Lyu, and I. King. "Improving recommender systems by incorporating social contextual information." ACM Transactions on Information Systems (TOIS), 29(2), 2011.
[17] J. Golbeck. "Computing with social trust." Human-Computer Interaction Series, Springer, 2008.
[18] Y. Koren, and R. Bell. "Advances in collaborative filtering." Recommender Systems Handbook, Springer US, 2011.
[19] Y. Koren. "Factorization meets the neighborhood: a multifaceted collaborative filtering model." In Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2008.
[20] Y. Koren, R. Bell, and C. Volinsky. "Matrix factorization techniques for recommender systems." Computer, 42(8): 30-37, 2009.
[21] S. Maniu, B. Cautis, and T. Abdessalem. "Building a signed network from interactions in Wikipedia." In Databases and Social Networks, 2011.
[22] V. Kant, and K. K. Bharadwaj. "Fuzzy computational models of trust and distrust for enhanced recommendations." International Journal of Intelligent Systems, 28(4): 332-365, 2013.
[23] C. C. Lin, T. S. Lin, and W. Y. Liu. "A trust and distrust mechanism for a social network-based recommendation system." In Wireless Personal Multimedia Communications (WPMC), 2012 15th International Symposium on. IEEE, 2012.
[24] B. Hajian. "On Measuring Influence and its Properties Social Networks." Diss. Carleton University, 2011.
[25] K. Naimisha. "Applications of social network analysis to community dynamics." February, MS Thesis, Indian Institute of Science, Bangalore, India, 2008.
[26] Z. Zhang, X. M. Wang, and Y. X. Wang. "A P2P global trust model based on recommendation." Machine Learning and Cybernetics, 2005. In Proceedings of 2005 International Conference on. Vol. 7. IEEE, 2005.
[27] M. Jamali, and M. Ester. "A matrix factorization technique with trust propagation for recommendation in social networks." In Proceedings of the fourth ACM Conference on Recommender Systems, 2010.
[28] X. Yang, H. Steck, Y. Guo, and Y. Liu. "On top-k recommendation using social networks." In Proceedings of the sixth ACM Conference on Recommender Systems, 2012.
[29] M. Jamali, and M. Ester. "A transitivity aware matrix factorization model for recommendation in social networks." In Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence(IJCAI), Vol. 11, 2011.
[30] S. H. Yang, A. J. Smola, B. Long, H. Zha, and Y. Chang. "Friend or frenemy?: predicting signed ties in social networks." In Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2012.
[31] C. N. Ziegler, and G. Lausen. "Spreading activation models for trust propagation." In Proceedings of the IEEE International Conference on e-Technology, e-Commerce, and e-Service (EEE '04), 2004.
[32] W. Dou, H. M. Wang, Y. Jia, and P. Zou. "A recommendation-based peer-to-peer trust model." Journal of Software, 15(4): 571-583, 2004.
[33] J. Golbeck. "Generating predictive movie recommendations from trust in social networks." In Proceedings of the fourth International Conference on Trust Management, 2006.
[34] Y. Kim, and H. S. Song. "Strategies for predicting local trust based on trust propagation in social networks." Knowledge-Based Systems, 24(8): 1360-1371, 2011.
[35] D. Artz, and Y. Gil. "A survey of trust in computer science and the semantic web." Web Semantics: Science, Services and Agents on the World Wide Web, 5(2): 58-71, 2007.
[36] W. Yuan, D. Guan, Y. K. Lee, S. Lee, and S. J. Hur. "Improved trust-aware recommender system using small-worldness of trust networks." Knowledge-Based Systems, 23(3): 232-238, 2010.
[37] S. Shekarpour, and S. D. Katebi. "Modeling and evaluation of trust with an extension in semantic web." Web Semantics: Science, Services and Agents on the World Wide Web, 8(1): 26-36, 2010.
[38] H. N. Kim, A. T. Ji, I. Ha, and G. S. Jo. "Collaborative filtering based on collaborative tagging for enhancing the quality of recommendation." Electronic Commerce Research and Applications, 9(1): 73-83, 2010.
[39] D. R. Liu, C. H. Lai, and W. J. Lee. "A hybrid of sequential rules and collaborative filtering for product recommendation." Information Sciences, 179(20): 3505-3519, 2009.
[40] Y. Y. Shih, and D. R. Liu. "Product recommendation approaches: Collaborative filtering via customer lifetime value and customer demands." Expert Systems with Applications, 35(1): 350-360, 2008.
[41] J. Bobadilla, F. Ortega, A. Hernando, and A. Gutierrez. "Recommender systems survey." Knowledge-Based Systems, 46: 109-132, 2013.
[42] N. Good, J. B. Schafer, J. A. Konstan, A. Borchers, B. Sarwar, J. Herlocker, and J. Riedl. "Combining collaborative filtering with personal agents for better recommendations." In Proceedings of the sixteenth national Conference on Artificial Intelligence and the eleventh Innovative Applications of Artificial Intelligence Conference Innovative Applications of Artificial Intelligence, 1999.
[43] A. G. Schick, L. A. Gordon, and S. Haka. "Information overload: A temporal approach." Accounting, Organizations and Society, 15(3): 199-220, 1990.
[44] C. N. Ziegler, and J. Golbeck. "Investigating interactions of trust and interest similarity." Decision Support Systems, 43(2): 460-475, 2007.
[45] J. Golbeck , and J. Hendler. "Computing and applying trust in web-based social networks." University of Maryland at College Park, College Park, MD, 2005.
[46] F. Ortega, J. A. Troyano, F. L. Cruz, C. G. Vallejo, and F. Enriquez. "Propagation of trust and distrust for the detection of trolls in a social network." Computer Networks, 56(12): 2884-2895, 2012.
[47] Y. Kim, and M. A. Ahmad. "Trust, distrust and lack of confidence of users in online social media-sharing communities." Knowledge-Based Systems, 37: 438-450, 2013.
[48] T. DuBois, J. Golbeck, and A. Srinivasan. "Predicting trust and distrust in social networks." In Proceedings of the third IEEE International Conference on Social Computing, 2011.