近年來因電子商務的興起，人們的消費型態也從傳統的實體店面購物，轉變為在網路商店上進行消費，推薦系統在電子商務上的運用很多，也有一定之成效，但對於實體零售業(Retail)而言，由於商家對於顧客消費並沒有設定門檻，無論是不是會員都可以進行商品購買，所以，商家只能針對會員的消費行為進行分析，但對於非會員的部份，卻無法有效掌握。
本研究是將推薦系統的概念應用在實體零售業上，運用協同過濾最常使用的方法Item-based進行研究，評估推薦系統套用在商品小類推薦之可行性。我們使用某知名企業所提供之資料作業實驗資料集，透過實驗分析結果，驗證本研究將小類別進行推薦是可行的，我們希望透過推薦系統的幫助，能增加商品的銷售，使消費者產生依賴及提高忠誠度，減少顧客流失。對廠商而言，也能更有效掌握消費者所購買的類別，並透過共同行銷、資訊交叉運用、產品組合等行銷策略，增加消費者下一次購買的可能性，吸引與留住顧客，提高市場佔有率。
因此，本研究的推薦方法能夠提供零售業做為行銷分析的基礎，轉換消費型態與思維，使會員能確切的獲得合適的推薦，超市進而增加營運與獲利。

Due to the rise of e-commerce in recent years, people’s consumption patterns have also changed from traditional physical storefront shopping to online shopping consumption. However, TRhe recommender system has a lot of applications in e-commerce and has also had some measure of success, but in physical retail, business firms never set the threshold for consumer spending and disregard whether or not members are able to purchase goods. Thus, business firms can only analyze members’ consumer behavior, but as non-members cannot be effectively controlled. 

This research on the conceptual application of the recommender system in physical retail was conducted using the item-based method often seen in collaborative filtering, and the feasibility of indiscriminately applying the recommender system in the small categories of recommended commodities was evaluated. We use a well-known enterprise data operations provided experimental data sets,it is hoped that the sales of the commodities can be strengthened with the help of the recommender system, thereby generating consumers’ reliance, improving the degree of loyalty, and reducing the loss o f customers. In terms of the manufacturers, all the categories of consumers’ purchases can also be even more effectively grasped, and the common marketing, information cross-application, product integration, and other marketing strategies will increase consumers’ subsequent purchase possibilities, attract and keep customers, and increase the market share.

Therefore, the recommendation methods of this research can adequately provide retail a basis for marketing analysis, transform consumption patterns and thinking, and enable members to exactly obtain the appropriate recommendations, thus strengthening supermarkets’ operations and profits.

目錄
第一章	緒論	1
1.1.	研究動機與目的	1
1.2.	論文架構	4
第二章	文獻探討	5
2.1.	推薦系統	5
2.2.	協同過濾	8
2.2.1.	協同過濾推薦技術	9
2.2.2.	協同過濾推薦步驟	10
2.2.3.	協同過濾應用	14
第三章	研究方法	18
3.1.	問題陳述	18
3.2.	研究設計	20
3.2.1.	研究架構	20
3.2.2.	系統推薦流程	21
第四章	研究結果分析	24
4.1.	資料介紹與預處理	24
4.2.	Item-based 協同過濾可行性驗證	28
第五章	結論與建議	34
參考文獻	35
附錄一 	39
附錄二 英文論文	47
 
圖目錄
圖 1、AMAZON網頁推薦系統	9
圖 2、協同過濾作業流程(Sarwar et al., 2001)	12
圖 3、研究架構圖	21
圖 4、Item-based運作流程	21
圖 5、推薦類別準確率趨勢圖	30
 
表目錄
表 1、推薦系統與搜索引擎比較表	6
表 2、商品小類別分類表	25
表 3、2011年4月至2012年1月推薦命中率	29
表 4、2011年4月新推薦1個小類別的購買率	33
表 5、4月新推薦2個小類別的購買率	39
表 6、4月新推薦3個小類別的購買率	40
表 7、5月新推薦1個小類別的購買率	41
表 8、5月新推薦2個小類別的購買率	42
表 9、5月新推薦3個小類別的購買率	43
表 10、6月新推薦1個小類別的購買率	44
表 11、6月新推薦2個小類別的購買率	45
表 12、6月新推薦3個小類別的購買率	46

Adomavicius, Gediminas, & Tuzhilin, Alexander. (2005). Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions. IEEE Trans. on Knowl. and Data Eng., 17(6), 734-749. doi: 10.1109/tkde.2005.99
Bell, Robert M., & Koren, Yehuda. (2007). Lessons from the Netflix prize challenge. SIGKDD Explor. Newsl., 9(2), 75-79. doi: 10.1145/1345448.1345465
Breese, John S., Heckerman, David, & Kadie, Carl. (1998). Empirical analysis of predictive algorithms for collaborative filtering. Paper presented at the Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence, Madison, Wisconsin. 
Burke, Robin. (2002). Hybrid Recommender Systems: Survey and Experiments. User Modeling and User-Adapted Interaction, 12(4), 331-370. doi: 10.1023/a:1021240730564
Candillier, Laurent, Meyer, Frank, & Boull, Marc. (2007). Comparing State-of-the-Art Collaborative Filtering Systems. Paper presented at the Proceedings of the 5th international conference on Machine Learning and Data Mining in Pattern Recognition, Leipzig, Germany. 
Deshpande, Mukund, & Karypis, George. (2004). Item-based top-N recommendation algorithms. ACM Trans. Inf. Syst., 22(1), 143-177. doi: 10.1145/963770.963776
Goldberg, David, Nichols, David, Oki, Brian M., & Terry, Douglas. (1992). Using collaborative filtering to weave an information tapestry. Commun. ACM, 35(12), 61-70. doi: 10.1145/138859.138867
Goldberg, Ken, Roeder, Theresa, Gupta, Dhruv, & Perkins, Chris. (2001). Eigentaste: A Constant Time Collaborative Filtering Algorithm. Inf. Retr., 4(2), 133-151. doi: 10.1023/a:1011419012209
Herlocker, Jonathan L., Konstan, Joseph A., & Riedl, John. (2000). Explaining collaborative filtering recommendations. Paper presented at the Proceedings of the 2000 ACM conference on Computer supported cooperative work, Philadelphia, Pennsylvania, United States. 
Kangning, Wei, Jinghua, Huang, & Shaohong, Fu. (2007, 9-11 June 2007). A Survey of E-Commerce Recommender Systems. Paper presented at the Service Systems and Service Management, 2007 International Conference on.
Konstan, Joseph A., Miller, Bradley N., Maltz, David, Herlocker, Jonathan L., Gordon, Lee R., & Riedl, John. (1997). GroupLens: applying collaborative filtering to Usenet news. Commun. ACM, 40(3), 77-87. doi: 10.1145/245108.245126
Lemire, D., & Maclachlan, A. (2005). Slope one predictors for online rating-based collaborative filtering. Proceedings of SIAM Data Mining (SDM'05).
Linden, G., Smith, B., & York, J. (2003). Amazon.com recommendations: item-to-item collaborative filtering. Internet Computing, IEEE, 7(1), 76-80. doi: 10.1109/MIC.2003.1167344
Pradel, Bruno, Sean, Savaneary, Delporte, Julien, Guerif, Sebastien, Rouveirol, Celine, Usunier, Nicolas, . . . Dufau-Joel, Frederic. (2011). A case study in a recommender system based on purchase data. Paper presented at the Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining, San Diego, California, USA. http://dl.acm.org/citation.cfm?id=2020470
Resnick, Paul, & Varian, Hal R. (1997). Recommender systems. Commun. ACM, 40(3), 56-58. doi: 10.1145/245108.245121
Rich, Elaine. (1979). User modeling via stereotypes. Cognitive Science, 3(4), 329-354. doi: http://dx.doi.org/10.1016/S0364-0213(79)80012-9
Sarwar, Badrul, Karypis, George, Konstan, Joseph, & Riedl, John. (2001). Item-based collaborative filtering recommendation algorithms. Paper presented at the Proceedings of the 10th international conference on World Wide Web, Hong Kong, Hong Kong. 
Schafer, J. Ben, Konstan, Joseph, & Riedi, John. (1999). Recommender systems in e-commerce. Paper presented at the Proceedings of the 1st ACM conference on Electronic commerce, Denver, Colorado, United States. 
Shardanand, Upendra, & Maes, Pattie. (1995). Social information filtering: algorithms for automating “word of mouth&rdquo. Paper presented at the Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Denver, Colorado, United States. 
Su, Xiaoyuan, & Khoshgoftaar, Taghi M. (2009). A survey of collaborative filtering techniques. Adv. in Artif. Intell., 2009, 2-2. doi: 10.1155/2009/421425
Terveen, L., & Hill, W. (2001). Beyond Recommender Systems: Helping People Help Each Other: Addison Wesley.
Wang, Jian, Sarwar, Badrul, & Sundaresan, Neel. (2011). Utilizing related products for post-purchase recommendation in e-commerce. Paper presented at the Proceedings of the fifth ACM conference on Recommender systems, Chicago, Illinois, USA.