隨著網際網路的使用者數量增加，購買和銷售活動從傳統管道轉移到網際網路上的交易或電子商務管道，這種情況讓企業更加重視收集顧客的訊息，但也導致訊息超載問題，因此，為了克服它，許多公司使用推薦系統來進行客製化，除了提供個人化訊息，並透過顧客的偏好來建議產品。本研究使用限制波茲曼機器(CRBM)進行協同過濾的分群，以梯度進化演算法(GE)最佳化限制波茲曼機器中的參數與梯度進化演算法的參數。此外，本研究也介紹了梯度進化演算法之向量跳躍法的新跳躍策略，其中包含高斯跳躍和柯西跳躍策略。
為了驗證本研究所提出的梯度進化演算法(GE)，本研究對每個資料集進行K-fold交叉驗證，實驗中使用了三個資料集包含：Movielens 100K、Book-Crossing和Restaurant。實驗證明，GE-CRBM可以優於結合其他萬用演算法的限制波茲曼機器(CRBM)，而所提出的高斯跳梯度進化演算法在大多數的資料集中，優於原始的梯度進化演算法(GE)。此外，不同的跳躍策略對不同的資料特性，具有不同的影響。

The increased number of internet users shifts the buying and selling activities from traditional channel to internet based transaction channel, or e-commerce channel. This situation allows companies to gather information as much as they can from customer. However, it causes an information-overloading problem. To overcome it, many companies use a recommendation system for customer personalization. The recommendation system provides customized information and suggests the product, using customer’s previous preferences. This study combine cluster based restricted Boltzman machine (CRBM) as collaborative filtering and gradient evolution (GE) algorithm to optimize both RBM’s parameters and GE’s parameters. Furthermore, this study also introduced new jumping strategies for vector jumping operator of GE algorithm including Gaussian jumping and Cauchy jumping strategies.
In order to validate the proposed GE algorithm, this study conducts K-fold cross validation for each dataset. Three datasets are used in the experiment including Movielens 100K, Book-Crossing, and Restaurant. The experiment result indicates that GE-CRBM with Gaussian jumping can outperform other metaheuristic cluster based-CRBM in most of datasets. In addition, each jumping strategy has different impact depending on the data characteristics.

Montaner, M., B. López, and J.L. De La Rosa, A taxonomy of recommender agents on the internet. Artificial intelligence review, 2003. 19(4): p. 285-330.
2. Resnick, P. and H.R. Varian, Recommender systems. Communications of the ACM, 1997. 40(3): p. 56-58.
3. Georgiev, K. and P. Nakov. A non-IID Framework for Collaborative Filtering with Restricted Boltzmann Machines. in ICML (3). 2013.
4. Su, X. and T.M. Khoshgoftaar, A survey of collaborative filtering techniques. Advances in artificial intelligence, 2009. 2009: p. 4.
5. Salakhutdinov, R., A. Mnih, and G. Hinton. Restricted Boltzmann machines for collaborative filtering. in Proceedings of the 24th international conference on Machine learning. 2007. ACM.
6. Chen, J.-T., An Application of differential evolution algorithm-based restricted Boltzmann machine to recommendation systems. 2016.
7. Tang, H., et al. Restricted Boltzmann machine based algorithm for multi-objective optimization. in Evolutionary Computation (CEC), 2010 IEEE Congress on. 2010. IEEE.
8. Hinton, G., A practical guide to training restricted Boltzmann machines. Momentum, 2010. 9(1): p. 926.
9. Kuo, R. and F.E. Zulvia, The gradient evolution algorithm: A new metaheuristic. Information Sciences, 2015. 316: p. 246-265.
10. Dong, H., et al., Evolutionary programming using a mixed mutation strategy. Information Sciences, 2007. 177(1): p. 312-327.
11. Sarwar, B., et al. Item-based collaborative filtering recommendation algorithms. in Proceedings of the 10th international conference on World Wide Web. 2001. ACM.
12. Cao, Y. and Y. Li, An intelligent fuzzy-based recommendation system for consumer electronic products. Expert Systems with Applications, 2007. 33(1): p. 230-240.
13. Lee, W.-P., C.-H. Liu, and C.-C. Lu, Intelligent agent-based systems for personalized recommendations in Internet commerce. Expert Systems with Applications, 2002. 22(4): p. 275-284.
14. Adomavicius, G. and A. Tuzhilin, Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. IEEE transactions on knowledge and data engineering, 2005. 17(6): p. 734-749.
15. Balabanović, M. and Y. Shoham, Fab: content-based, collaborative recommendation. Communications of the ACM, 1997. 40(3): p. 66-72.
16. Pazzani, M. and D. Billsus, Content-based recommendation systems. The adaptive web, 2007: p. 325-341.
17. Lops, P., M. De Gemmis, and G. Semeraro, Content-based recommender systems: State of the art and trends, in Recommender systems handbook. 2011, Springer. p. 73-105.
18. Burke, R., Hybrid web recommender systems. The adaptive web, 2007: p. 377-408.
19. Vincent, P., et al. Extracting and composing robust features with denoising autoencoders. in Proceedings of the 25th international conference on Machine learning. 2008. ACM.
20. Dahl, G.E., et al., Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition. IEEE Transactions on Audio, Speech, and Language Processing, 2012. 20(1): p. 30-42.
21. Fischer, A. and C. Igel, Training restricted Boltzmann machines: An introduction. Pattern Recognition, 2014. 47(1): p. 25-39.
22. Hopfield, J.J., Neural networks and physical systems with emergent collective computational abilities. Proceedings of the national academy of sciences, 1982. 79(8): p. 2554-2558.
23. Hinton, G.E., Training products of experts by minimizing contrastive divergence. Neural computation, 2002. 14(8): p. 1771-1800.
24. Carreira-Perpinan, M.A. and G.E. Hinton. On Contrastive Divergence Learning. in AISTATS. 2005. Citeseer.
25. Jain, A.K., Data clustering: 50 years beyond K-means. Pattern recognition letters, 2010. 31(8): p. 651-666.
26. Kanungo, T., et al., An efficient k-means clustering algorithm: Analysis and implementation. IEEE transactions on pattern analysis and machine intelligence, 2002. 24(7): p. 881-892.
27. Bezdek, J.C., R. Ehrlich, and W. Full, FCM: The fuzzy c-means clustering algorithm. Computers & Geosciences, 1984. 10(2-3): p. 191-203.
28. Joshi, A. and R. Kaur, A review: Comparative study of various clustering techniques in data mining. International Journal of Advanced Research in Computer Science and Software Engineering, 2013. 3(3).
29. Talbi, E.-G., Metaheuristics: from design to implementation. Vol. 74. 2009: John Wiley & Sons.
30. Holland, J.H., Genetic algorithms. Scientific american, 1992. 267(1): p. 66-72.
31. Kennedy, J. and R. Eberhart. Particle swarm optimization (PSO). in Proc. IEEE International Conference on Neural Networks, Perth, Australia. 1995.
32. Storn, R. and K. Price, Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces. Journal of global optimization, 1997. 11(4): p. 341-359.
33. Krishna, K. and M.N. Murty, Genetic K-means algorithm. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 1999. 29(3): p. 433-439.
34. Michielssen, E., S. Ranjithan, and R. Mittra, Optimal multilayer filter design using real coded genetic algorithms. IEE Proceedings J (Optoelectronics), 1992. 139(6): p. 413-420.
35. Li, Z., Y. Li, and L. Xu. Anomaly intrusion detection method based on k-means clustering algorithm with particle swarm optimization. in Information Technology, Computer Engineering and Management Sciences (ICM), 2011 International Conference on. 2011. IEEE.