當使用者對商品的評分數量不足時，會導致了推薦系統沒有足夠的資訊可以利用，而無法有效發揮其作用，這種情況稱為冷起始問題。過去協同過濾技術是推薦系統中重要的一環，它以使用者為基礎或以商品為基礎的協同過濾方式找到足夠且相似的使用者或商品來做推薦。然而，此方法卻無法有效解決冷起始問題，冷起始是一個存在於以電腦為基礎涉及自動化數據建模之資訊系統的問題。更精確的來說，這個問題所碰觸到的議題是有關於一個系統完全無法對某些資訊量不足的使用者或商品做出任何推論。再者，現今有愈來愈多的網站開始提供使用者與使用者之間的關係，例如信任網路，我們就可以利用這項資訊幫助我們緩和前面的冷起始問題。在此論文中，我們提出了RSOL模型，這個模型可以辨識出每位使用者對於不同的商品的在推薦上的信心度。一位使用者的信心度包含兩個部分，Rating Confidence: 使用者對於一項商品評分的信心度; Proximity Prestige: 使用者在信任網路的影響力。最後，我們使用Epinions資料集來對我們提出的模型做了相關評估，並且將我們的模型與傳統的協同過濾方法和現今以使用者之間的信任度為基礎的方法做了比較。結果是，RSOL模型比現在的目前最先進的方法效能為佳，RSOL模型能較之前的方法有效解決冷起始問題。

Collaborative Filtering (CF) technique is the essential part of recommender systems. However, the Sparsity of the user item ratings makes the traditional CF methods failed. Due to the less of user item ratings, User-based or Item-based CF methods cannot find enough similar users or items to do the predictions. We call the situation cold start. The cold start problem is a potential issue in computer-based information systems that involve a degree of automated data modeling. Specifically, the system cannot infer a rating for users or items that are new to the recommeder system, when no sufficient information has been gather. Currently, more websites are providing the relationships between users, e.g., the trust relationships, to help us alleviate the cold start problem. In this paper, we proposed a trust-based recommender system model (RSOL) that is able to recognize the user’s recommendation quality for different items. A user quality contains two parts: “Rating Confidence”- an indicator of the user’s reliability when rating an item, and “Proximity Prestige”- an indicator of the user’s influence on the trust network. In our experimental results, the proposed method outperforms the existing Collaborative Filtering and trust-based methods on the Epinions dataset.

[1] Adomavicius, G., N. Manouselis, and Y. Kwon, Multi-Criteria Recommender Systems, in Recommender Systems Handbook, F. Ricci, et al., Editors. 2011, Springer US. p. 769-803.
[2] Alvarez-Hamelin, J., L. Dall'asta, A. Barrat, and A. Vespignani, k-core decomposition: a tool for the visualization of large scale networks. arXiv, 2005.
[3] Balabanovi, M., #263, and Y. Shoham, Fab: content-based, collaborative recommendation. Commun. ACM, 1997. 40(3): p. 66-72.
[4] Crandall, D., D. Cosley, D. Huttenlocher, J. Kleinberg, and S. Suri, Feedback effects between similarity and social influence in online communities, in Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. 2008, ACM: Las Vegas, Nevada, USA. p. 160-168.
[5] Gantner, Z., L. Drumond, C. Freudenthaler, S. Rendle, and L. Schmidt-Thieme, Learning Attribute-to-Feature Mappings for Cold-Start Recommendations, in Proceedings of the 2010 IEEE International Conference on Data Mining. 2010, IEEE Computer Society. p. 176-185.
[6] Golbeck, J.A., Computing and applying trust in web-based social networks. 2005, University of Maryland at College Park. p. 199.
[7] Jamali, M. 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, ACM: Barcelona, Spain. p. 135-142.
[8] Jamali, M. and M. Ester, TrustWalker: a random walk model for combining trust-based and item-based recommendation, in Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining. 2009, ACM: Paris, France. p. 397-406.
[9] Jung, M.K., M.K. Jang, H.K. Kim, and J.K. Kim, A New Item Recommendation Procedure Based on Min-Max Distance.
[10] Kim, J.K., M.K. Jang, H.K. Kim, and Y.H. Cho, A hybrid recommendation procedure for new items using preference boundary, in Proceedings of the 11th International Conference on Electronic Commerce. 2009, ACM: Taipei, Taiwan. p. 289-295.
[11] Koenigstein, N., G. Dror, and Y. Koren, Yahoo! music recommendations: modeling music ratings with temporal dynamics and item taxonomy, in Proceedings of the fifth ACM conference on Recommender systems. 2011, ACM: Chicago, Illinois, USA. p. 165-172.
[12] Koren, Y., 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, ACM: Las Vegas, Nevada, USA. p. 426-434.
[13] Koren, Y., R. Bell, and C. Volinsky, Matrix Factorization Techniques for Recommender Systems. Computer, 2009. 42(8): p. 30-37.
[14] Ma, H., H. Yang, M.R. Lyu, and I. King, SoRec: social recommendation using probabilistic matrix factorization, in Proceedings of the 17th ACM conference on Information and knowledge management. 2008, ACM: Napa Valley, California, USA. p. 931-940.
[15] Ma, H., D. Zhou, C. Liu, M.R. Lyu, and I. King, Recommender systems with social regularization, in Proceedings of the fourth ACM international conference on Web search and data mining. 2011, ACM: Hong Kong, China. p. 287-296.
[16] Massa, P. and P. Avesani. Trust-aware bootstrapping of recommender systems. in ECAI Workshop on Recommender Systems. 2006: Citeseer.
[17] Massa, P. and P. Avesani, Trust-aware recommender systems, in Proceedings of the 2007 ACM conference on Recommender systems. 2007, ACM: Minneapolis, MN, USA. p. 17-24.
[18] Mooney, R.J. and L. Roy, Content-based book recommending using learning for text categorization, in Proceedings of the fifth ACM conference on Digital libraries. 2000, ACM: San Antonio, Texas, USA. p. 195-204.
[19] Pazzani, M. and D. Billsus, Learning and Revising User Profiles: The Identification of Interesting Web Sites. Machine Learning, 1997. 27(3): p. 313-331.
[20] Sarwar, B., G. Karypis, J. Konstan, and J. Riedl, Item-based collaborative filtering recommendation algorithms, in Proceedings of the 10th international conference on World Wide Web. 2001, ACM: Hong Kong, Hong Kong. p. 285-295.
[21] Schein, A.I., A. Popescul, L.H. Ungar, and D.M. Pennock, Methods and metrics for cold-start recommendations, in Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval. 2002, ACM: Tampere, Finland. p. 253-260.
[22] Wang, J., A.P.d. Vries, and M.J.T. Reinders, Unifying user-based and item-based collaborative filtering approaches by similarity fusion, in Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval. 2006, ACM: Seattle, Washington, USA. p. 501-508.
[23] Xu, W., X. Liu, and Y. Gong, Document clustering based on non-negative matrix factorization, in Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval. 2003, ACM: Toronto, Canada. p. 267-273.