跨領域協同過濾利用來自多個領域的評分矩陣，以做出更好的推薦結果。現有的跨領域協同過濾方法採用子結構共享技術，只能在各領域之間傳送線性相關的資訊知識。在本文中，我們提出了超結構遷移的概念，即要求每個評分矩陣由一個所有領域共有的更複雜的結構的投影部分進行說明，稱為超結構，因此可以讓所有領域之間的非線性相關的知識得以鑑識和轉移。並以充實的實驗方式進行驗證，結果可以顯示我們的超領域遷移模型具有有效性。

The Cross Domain Collaborative Filtering (CDCF) exploits the rating matrices from multiple domains to make better recommendations. Existing CDCF methods adopt the sub-structure sharing technique that can only transfer linearly correlated knowledge between domains. In this paper, we propose the notion of Hyper-Structure Transfer (HST) that requires the rating matrices to be explained by the projections of some more complex structure, called the hyper-structure, shared by all domains, and thus allows the non-linearly correlated knowledge between domains to be identified and transferred. Extensive experiments are conducted and the results demonstrate the effectiveness of our HST models empirically.

Table of Contents
透過超結構遷移的非線性跨領域協同過濾 ii
Non-Linear Cross-Domain Collaborative Filtering via Hyper-Structure Transfer iii
Acknowledgments iv
Abstract v
摘要 vi
Table of Contents vii
List of Figures ix
List of Tables x
Chapter 1 Introduction 1
Chapter 2 Challenges and Evidence 6
2.1. Linear Knowledge Transfer 6
2.2. Maladaptive Bridge 7
Chapter 3 Hyper-Structure Transfer 9
3.1. General Model 9
3.2. MOTAR 10
3.3. Residuals for Adaptive Bridge 11
Chapter 4 Objective Solving 12
Chapter 5 Experiment 15
5.1. Results over Two Domains 16
5.2. Results over Multiple Domains 17
5.3. Effects of Non-linearity 18
Chapter 6 Conclusions and Future Work 19
Bibliography 21

Bibliography
[1] Argyriou, Andreas, Andreas Maurer, and Massimiliano Pontil. An algorithm or transfer learning in a heterogeneous environment. In proc. of ECML PKDD, pages 71-85. ECML PKDD, 2008.
[2] Bader, Brett W. and Kolda, Tamara G. Algorithm 862:MATLAB tensor classes for fast algorithm prototyping. ACM Trans. on Mathematical Software, 32:635–653, 2006.
[3] Bakker, Bart and Heskes, Tom. Task clustering and gating for bayesian multitask learning. The Journal of Machine Learning Research, 4:83–99, 2003.
[4] Ben-David, S. and Schuller, R. Exploiting task relatedness for multiple task learning. In Proc. of COLT/Kernel, 2003.
[5] Cao, Bin, Liu, Nathan N, and Yang, Qiang. Transfer learning for collective link prediction in multiple heterogenous domains. In Proc. of ICML, 2010.
[6] Chen, Wei, Hsu, Wynne, and Lee, Mong Li. Making recommendations from multiple domains. In Proc. of KDD, 2013.
[7] Ding, Chris, Li, Tao, Peng, Wei, and Park, Haesun. Orthogonal nonnegative matrix t-factorizations for clustering. In Proc. of KDD, 2006.
[8] Gao, Sheng, Luo, Hao, Chen, Da, Li, Shantao, Gallinari, Patrick, and Guo, Jun. Cross-domain recommendation via cluster-level latent factor model. In Proc. of ECML PKDD, 2013a.
[9] Gao, Sheng, Luo, Hao, Chen, Da, Li, Shantao, Gallinari, Patrick, Ma, Zhanyu, and Guo, Jun. A cross-domain recommendation model for cyber-physical systems. IEEE Trans. on Emerging Topics in Computing, 1:384–393, 2013b.
[10] Gu, Quanquan, Ding, Chris, and Han, Jiawei. On trivial solution and scale transfer problems in graph regularized nmf. In Proc. of IJCAI, 2011.
[11] Hofmann, Thomas. Latent semantic models for collaborative filtering. ACM Trans. on Information Systems, 22:89–115, 2004.
[12] Hu, Liang, Cao, Jian, Xu, Guandong, Cao, Longbing, Gu, Zhiping, and Zhu, Can. Personalized recommendation via cross-domain triadic factorization. In Proc. of WWW, 2013.
[13] Hu, Yifan, Koren, Yehuda, and Volinsky, Chris. Collaborative filtering for implicit feedback datasets. In Proc. of ICDM, 2008.
[14] Kiers, Henk AL, Ten Berge, Jos MF, and Bro, Rasmus. Parafac2-part i. a direct fitting algorithm for the parafac2 model. Journal of Chemometrics, 13:275–294, 1999.
[15] Kolda, Tamara G and Bader, Brett W. Tensor decompositions and applications. SIAM review, 51:455–500, 2009.
[16] Koren, Yehuda, Bell, Robert, and Volinsky, Chris. Matrix factorization techniques for recommender systems. Computer, 42:30–37, 2009.
[17] Li, Bin. Cross-domain collaborative filtering: A brief survey. In Proc. of ICTAI, 2011.
[18] Li, Bin, Yang, Qiang, and Xue, Xiangyang. Can movies and books collaborate? cross-domain collaborative filtering for sparsity reduction. In Proc. of IJCAI, 2009.
[19] Long, Mingsheng, Wang, Jianmin, Ding, Guiguang, Shen, Dou, and Yang, Qiang. Transfer learning with graph co-regularization. In Proc. of AAAI, 2012.
[20] Moreno, Orly, Shapira, Bracha, Rokach, Lior, and Shani,Guy. Talmud: transfer learning for multiple domains. In Proc. of CIKM, 2012.
[21] Pan, Weike, Xiang, Evan Wei, Liu, Nathan Nan, and Yang, Qiang. Transfer learning in collaborative filtering for sparsity reduction. In Proc. of AAAI, 2010.
[22] Rosenstein, Michael T, Marx, Zvika, Kaelbling, Leslie Pack, and Dietterich, Thomas G. To transfer or not to transfer. In Proc. of NIPS, 2005.
[23] Schölkopf, Bernhard and Smola, Alexander J. Learning with kernels: support vector machines, regularization, optimization, and beyond. MIT press, 2002.
[24] Shi, Yue, Larson, Martha, and Hanjalic, Alan. Tags as bridges between domains: Improving recommendation with tag-induced cross-domain collaborative filtering. User Modeling, Adaption and Personalization, 6787:305–316, 2011.
[25] Su, Xiaoyuan and Khoshgoftaar, Taghi M. A survey of collaborative filtering techniques. Advances in artificial intelligence, 2009:4, 2009.
[26] Tang, Jie, Zhang, Jing, Yao, Limin, Li, Juanzi, Zhang, Li, and Su, Zhong. Arnetminer: extraction and mining of academic social networks. In Proc. of KDD, 2008.
[27] Zhang, Yu, Cao, Bin, and Yeung, Dit-Yan. Multi-domain collaborative filtering. arXiv preprint arXiv:1203.3535, 2012.