目前大多數電子商務網站使用的推薦系統(Recommender System)，多為協同過濾(Collaborative Filtering, CF)演算法之Memory-based模式，其優點是能夠快速實現商品推薦之功能，但缺點為1. 當資料庫增大時，因複雜度太高(Complexity, O(g(n)))而無法擴展(scalability)的問題，造成推薦的效能不佳；2. 此演算法仰賴用戶對商品的評價，故當用戶成功交易後，若無給予商品評價，或因用戶和商品大量時，皆會因評價數不足造成數據稀疏性(Data Sparsity)，因而影響鑑別商品之推薦度問題；3. 而另一個因數據稀疏性引申的問題，則是在新用戶和新商品加入時，因無法判定其用戶喜好或用戶對商品的評價，也會造成冷啟始(Cold-start)狀況。
為改進以上之缺陷，本研究中針對現有的技術與文獻進行探討後，提出一C2C(customers to customers)購物網站-IMI Lab Global，提供用戶販售與購買商品，針對Memory-based的缺點，提出三大解決方法：1. 為因應用戶與商品巨量資料(Big Data)的擴展性問題，本研究除了RDB關聯性資料庫，也使用Apache開放原始碼之Hadoop環境，以HDFS架構處理NoSQL資料，即可加速巨量資料的處理速度，實現高吞吐量和低延遲性；2. 為減少因用戶不願或忘記評價導致評價數不足，亦即數據稀疏性的問題，本研究提出一個時間過濾評價(Rating time-filter)方法，補足用戶購買後無評價造成的缺陷；3. 在新用戶和新商品加入時，造成的冷啟始問題。本研究提出之方法，可結合社群網路(Social Network)，提供用戶多樣的登入口，並除註冊會員資料外不同面向提取用戶資訊，來做為用戶之背景資料分析。

Recommender System is implementing in so many e-commerce websites like Amazon and Taobao which are the most famous C2C websites all over the world, and the main algorithm they used is Collaborative Filtering (CF), which could realize the recommender system quickly since the parameters are few. However, the disadvantages are still there, like Scalability problem, Data Spartsity problem, and Cold-start problem, which are discussing in most thesis in the past.

To improve the Recommender efficiency, this research implement a C2C shopping website - ”IMI Lab Global” for customers to sell and also purchase products, and there are 3 solutions to solve the problems that CF algorithm have, 1. for scalability problem, this research not only use the RDB but Apache Hadoop environment to deal with NoSQL data to realize the processing speed, high throughput and low latency, 2. for data sparsity problem, this research gives a Rating time-filter solution to give ratings on the user-item matrix which we will use for CF memory-based model, 3. for Cold-start problem, this research uses Facebook access for customers to login into the shopping website ”IMI Lab Global”, and then we may take the user’s preference as our user’s similarity concept.

[1] Wikipedia, “E-commerce,” Internet: http://en.wikipedia.org/wiki/E-commerce, Jul. 2010 [Jan. 24, 2014].
[2] Amazon Inc. “Amazon.com,” Internet: http://www.amazon.com, May. 2012 [Jan. 24, 2014].
[3] Netflix Inc. “Netflix.com,” Internet: https://www.netflix.com/global, Jun. 2011 [Jan. 24, 2014].
[4] Taobao.com, “淘寶網,” Internet: http://tw.taobao.com, Sep. 2013 [Jan. 14, 2014].
[5] 博客來數位科技股份有限公司, “博客來,” Internet: http://www.books.com.tw, Jul. 2013 [Oct. 24, 2013].
[6] Google AdWords, Internet: https://support.google.com/adwords/answer/2684489?hl=zh-Hant, Nov. 2013 [Feb. 4, 2014].
[7] Chih-Fong Tsai, Chihli Hung, “Cluster ensembles in collaborative filtering recommendation,” Applied Soft Computing, Vol. 12, Issue 4, pp. 1417-1425, 2012.
[8] Manuela I. Martín-Vicente, Alberto Gil-Solla, Manuel Ramos-Cabrer, Yolanda Blanco-Fernández, Martín López-Nores, “Semantic inference of user’s reputation and expertise to improve collaborative recommendations,” Expert Systems with Applications, Vol. 39, Issue 9, pp. 8248-8258, Jul. 2012.
[9] StackDOC, “如何提高網站的信任度” Internet: http://rritw.com/a/caozuoxitong/android/2011/1122/145336.html, Jul. 2009 [Apr. 24, 2013].
[10] Wikipedia, “Collaborative filtering,” Internet: http://en.wikipedia.org/wiki/Collaborative_filtering, Jul. 2013 [Jan. 22, 2014].
[11] Wikipedia, “長尾效應,” Internet: http://en.wikipedia.org/wiki/Long_tail, Jul. 2013 [Jan. 24, 2014].
[12] 美國風險投資公司, “Kleiner Perkins Caufield & Byers,” Internet: http://www.kpcb.com, Feb. 2013 [Jan. 17, 2014].
[13] 阿里巴巴集團, ”Alibaba Group,” Internet: http://www.alibabagroup.com/cn/global/home, Jul. 2010 [Jan. 20, 2014].
[14] Google AdWords, “不重複訪客/活躍使用者的計算方式,” Internet: https://support.google.com/analytics/answer/2992042?hl=zh-Hant, Dec. 2010 [Jun. 3, 2014].
[15] About.com Useconomy, “Year-Over-Year Growth Rate,” Internet: http://useconomy.about.com/od/glossary/g/year_over_year.htm, Jul. 2013 [Jan. 8, 2014].
[16] WSJ, “What is Alibaba,” Internet: http://projects.wsj.com/alibaba/?mod=e2tw#chapter4, Nov. 2013 [Jan. 2, 2014].
[17] Wikipedia, “Gross merchandise volume,” Internet: http://en.wikipedia.org/wiki/Gross_merchandise_volume, Oct. 2013 [Jan. 16, 2014].
[18] eBay, “eBay,” Internet: http://www.ebay.com.tw, Feb. 2013 [Jan. 25, 2014].
[19] 31會議網, “阿里技術嘉年華,” Internet: http://www.31huiyi.com/page/2515, Mar. 2013 [Jan. 17, 2014].
[20] 人人IT網, “淘寶推薦系統,” http://rritw.com/a/bianchengyuyan/C__/20130609/367418.html, Aug. 2013 [Jan. 24, 2014].
[21] Linyuan Lü, Matúš Medo, Chi Ho Yeung, Yi-Cheng Zhang, Zi-Ke Zhang, Tao Zhou, “Recommender systems,” Physics Reports, Vol. 519, Issue 1, pp. 1-49, Oct. 2012.
[22] Jürgen Buder, Christina Schwind, “Learning with personalized recommender systems: A psychological view,” Computers in Human Behavior, Vol. 28, Issue 1, pp. 207-216, Jan. 2012.
[23] Fortune, “Amazon’s recommendation secret,” Internet: http://fortune.com/2012/07/30/amazons-recommendation-secret, Jul. 2013 [Jan. 24, 2014].
[24] Linden, G., Smith, B., & York, J., “Amazon.com recommendations: item-to-item collaborative filtering,” Internet Computing, IEEE, Vol. 7, Issue 1, pp. 76-80, 2003, doi:10.1109/MIC.2003.1167344.
[25] Zhihu, “Amazon 的推薦算法是否特別優秀,” Internet: http://www.zhihu.com/question/19969010, Nov. 2008 [Jan. 16, 2014].
[26] Amazon Web Services, “e-commerce API,” Internet: http://aws.amazon.com/cn/ecommerce-applications, Jun. 2006 [May. 9, 2014].
[27] Dongsheng Li, Qin Lv, Li Shang, Ning Gu, “Item-based top-N recommendation resilient to aggregated information revelation,” Knowledge-Based Systems, Vol. 67, pp. 290-304, Sep. 2014.
[28] Paul Resnick, Neophytos Iacovou, Mitesh Suchak, Peter Bergstrom, John Riedl, “GroupLens: An Open Architecture for Collaborative Filtering of Netnews,” ACM Conference on Computer Supported Cooperative Work, Chapel Hill, pp. 175-186, 1994.
[29] Wikipedia, “GroupLens Research,” Internet: http://en.wikipedia.org/wiki/GroupLens_Research, Jul. 2013 [Jan. 24, 2014].
[30] Xiaoyuan Su, Taghi M. Khoshgoftaar, “A Survey of Collaborative Filtering Techniques,” Advances in Artificial Intelligence, Vol. 2009(2009), Article ID 421425, pp. 1-19, doi:10.1155/2009/421425.
[31] Wikipedia, “Pearson product-moment correlation coefficient,” Internet: http://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient, Feb. 2012 [Aug. 24, 2014].
[32] Jun Wang, Arjen P. de Vries, Marcel J. T. Reinders, “Unifying user-based and item-based collaborative filtering approaches by similarity fusion,” SIGIR '06 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval, pp. 501-508, 2006, doi:10.1145/1148170.1148257.
[33] Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl, “Item-Based Collaborative Filtering Recommendation Algorithms,” Hong Kong, ACM 1-58113-348-0/01/0005, pp. 285-295, 2001.
[34] Jun Wang, Arjen P. de Vries, Marcel J. T. Reinders, “Unifying user-based and item-based collaborative filtering approaches by similarity fusion,” SIGIR '06 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval, pp. 501-508, 2006, doi:10.1145/1148170.1148257.
[35] Chih-Fong Tsai, Chihli Hung, “Cluster ensembles in collaborative filtering recommendation,” Applied Soft Computing, Vol. 12, Issue 4, pp. 1417-1425, Apr. 2012.
[36] Guibing Guo, Jie Zhang, Daniel Thalmann, “Merging trust in collaborative filtering to alleviate data sparsity and cold start,” Knowledge-Based Systems, Vol. 57, pp. 57-68, Feb. 2014.
[37] Yehuda Koren, “The BellKor Solution to the Netflix Grand Prize,” Netflix prize documentation, Vol. 81, pp. 1-10, 2009.
[38] Jianming He , Wesley W. Chu, “A Social Network-Based Recommender System (SNRS),” In Annals of Information Systems, Special Issue on Data Mining for Social Network Data, Vol. 12, pp. 47-74, 2010, doi:10.1.1.156.2547.
[39] Xiwang Yang, Harald Steck, Yang Guo, Yong Liu, “On Top-k Recommendation using Social Networks,” RecSys '12 Proceedings of the sixth ACM conference on Recommender systems, pp. 67-74, 2012, doi:10.1145/2365952.2365969.
[40] Ioannis Konstas, Vassilios Stathopoulos, Joemon M. Jose, “On social networks and collaborative recommendation,” SIGIR '09 Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval, pp. 195-202, 2009, doi:10.1145/1571941.1571977.
[41] Fengkun Liu, Hong Joo Lee, “Use of social network information to enhance collaborative filtering performance,” Expert Systems with Applications, Vol. 37, Issue 7, pp. 4772–4778, 2010, doi:10.1016/j.eswa.2009.12.061.
[42] Facebook Inc. “Facebook.com,” Internet: http://www.facebook.com, Jul. 2012[Jan. 2, 2014].
[43] 數位時代, “Facebook 10周歲，全球用戶12.3億人,” Internet: http://www.bnext.com.tw/article/view/id/30952, May. 2013 [Jun. 13, 2014].
[44] iThome, “國內Facebook 有效用戶超過1500萬人”, Internet: http://www.ithome.com.tw/news/88943, Aug. 2013 [Jun. 8, 2014].
[45] Facebook Inc., “Facebook Developers,” Internet: https://developers.facebook.com/, Jun. 2008 [May. 2, 2014].
[46] Wikipedia, “Multitier architecture,” Internet: http://en.wikipedia.org/wiki/Multitier_architecture, Jul. 2013 [Jan. 2, 2014].
[47] The Apache Software Foundation, “Hadoop,” Internet: http://hadoop.apache.org/, Jun. 2009 [Mar. 9, 2014].
[48] The Apache Software Foundation, “HDFS,” Internet: http://hadoop.apache.org/docs /r1.2.1/hdfs_design.html, Jul. 2013 [Jun. 4, 2014].
[49] The Apache Software Foundation, “MapReduce,” Internet: http://hadoop.apache.org/docs/r1.2.1/mapred_tutorial.html, Jul. 2010 [Jan. 24, 2013].
[50] Makoto SHIROTA, “Big Data大數據的獲利模式：圖解．案例．策略．實戰,” 經濟新潮社, pp. 145-146, 2013.
[51] Netflix Inc., “System Architectures for Personalization and Recommendation,” Internet: http://techblog.netflix.com/2013/03/system-architectures-for.html, May. 2009 [Jan. 14, 2014].
[52] Francesco Ricci, Lior Rokach, Bracha Shapira, Paul B. Kantor, “Recommender Systems Handbook,” Springer, 2010.
[53] Wikipedia, “Online transaction processing,” Internet: http://en.wikipedia.org/wiki/Online_transaction_processing, Jul. 2012 [Jan. 24, 2014].
[54] 資策會產業情報研究所（MIC）, “2013年台灣網購行為-使用行動裝置消費額大幅成長,” Internet: http://mic.iii.org.tw/aisp/pressroom/press01_pop.asp?sno=353&type1=2, Aug. 2013 [Jun. 24, 2014].
[55] 朱郁筱, 呂琳媛, “推薦系統評價指標綜述,” 電子科技大學學報，中國，2012.