過去傳統產業(例如：紡織業、傢俱與寢具業等。)曾扮演過台灣經濟成長不可或缺的角色，在外銷的黃金年代帶領台灣經濟起飛，其年產量曾在全世界佔有一定地位，這讓當時台灣社會逐漸從農村經濟逐漸轉向工業化經濟，並為大量農村人口提供就業機會；但經歷過經濟蕭條、人力成本的上漲與中國經濟的崛起等因素，幾經波折後使台灣傳統產業逐漸失去過往國際上的外銷優勢，因而被迫轉型。
對於專注於實體店面銷售的傳統產業而言，轉型意味著行銷手段也必須與時俱進，由於現今較少研究人員特別對於傳統產業提出有利的行銷方案，基於此，本研究對於類似情形的傳統業者提出了一個可行的方案，利用行銷學上經典分析客戶價值的方⏤RFM，業者不需要複雜的交易資料集，只需要有每日的交易金額、交易日期與會員編號，就可對於每位顧客的價值做分析，整理好顧客RFM 特徵後針後搭配機器學習演算法中強大的預測分類方法(Random Forest Classifier)預測顧客的回購情形，最後運用探索顧客購買品項紀錄中知名的方法⏤先驗先算演算法(Apriori Algorithm)分析顧客可能想購買的品項，進而優化傳統產業的行銷結果。我們使用台灣某轉型成功的寢具廠商資料進行預測分析，研究結果顯示本研究的方法確實捕捉到顧客的購物偏好。

In the past, traditional industries (such as textiles, furniture, and bedding industries)have played an indispensable role in Taiwan’s economic growth, leading Taiwan’s economy to take off in the golden age of exports, which has led to a gradual shift from an agricultural to industrialized economy and provided many vacancies for a large rural population. However, after experiencing an economic depression, rising labor costs, and the rise of China's economy among other factors, Taiwan’s traditional industries have gradually lost their international export advantages and have been forced to transform.

For the traditional industries in Taiwan that focused on physical stores sales, transformation meant that marketing methods must also keep abreast of the times, Few modern scholars have proposed favorable marketing plans, especially for traditional industries. Therefore, this research based on the traditional industries has proposed a feasible solution, using the classic method of analyzing customer value in marketing :Recency,Frequency, Monetary (RFM). Sorting out the customer's RFM characteristics and combining the powerful predictive classification method called Random Forest Classifier in machine learning algorithms to predict customer's repurchase behavior, Finally, use the well-known method to explore customer purchase item records with the Apriori Algorithm to analyze what items customers may want to buy and then optimize the marketing results of traditional industries.

We use the data of a successful bedding retailer in Taiwan to conduct predictive analysis. The results of the research show that the method in this research indeed captures customer shopping preferences.

中文文獻
1. 蔡宜霖, 應用機器學習於零售通路顧客之回購機率預測 , (成功大學會計研
究所論文，研究進行中).

英文文獻
1. Allen, R. B.,“User models: theory, method, and practice”, International Journal of man-machine Studies, Vol.32(5), pp. 511-543, 1990.

2. Agrawal R., Imielinski T., A Swam, A.,“ Mining association rules between sets of items in large databases”, 1993 ACM SIGMOD international conference on Management of data , pp. 207-216, 1993.

3. Baesens, B., Viaene, S., Van den Poel, D., Vanthienen, J., & Dedene, G.,“Bayesian neural network learning for repeat purchase modelling in direct marketing”, European Journal of Operational Research, Vol.138(1), pp. 191-211, 2002.

4. Beel, J., Genzmehr, M., Langer, S., Nürnberger, A., & Gipp, B.,“A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation”, Paper presented at the In Proceedings of the international workshop on reproducibility and replication in recommender systems evaluation , pp. 7-14, 2013.

5. Bengio, Y., & Grandvalet, Y.,“No Unbiased Estimator of the Variance of KFold Cross-Validation”, The Journal of Machine Learning Research, Vol.5(Sep),pp. 1089-1105, 2004.

6. Bobadilla, J., Ortega, F., Hernando, A., & Bernal, J., “A collaborative filtering approach to mitigate the new user cold start problem”, Knowledge-Based Systems, 26, pp. 225-238, 2012.

7. Breiman, L., “Random forests”, Machine learning, Vol.45(1), pp. 5-32, 2001.

8. Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A.,“Classification and regression trees”, CRC press, Florida, United States , 1984.

9. Bult, J. R., & Wansbeek, T.,“Optimal selection for direct mail”, Marketing Science, Vol.14(4), pp. 378-394 , 1995.

10. Chen, M. S., Han, J., & Yu, P. S., “Data Mining An overview from a database perspective”, IEEE Transactions on Knowledge and data Engineering, Vol.8(6), pp. 866-883 , 1996.

11. Covington, P., Adams, J., & Sargin, E.,“Deep Neural Networks for YouTube Recommendations”, 10th ACM conference on recommender systems, pp. 191-198, 2016.

12. Fader, P. S., Hardie, B. G. S., & Lee, K. L.,“RFM and CLV: Using iso-value curves for customer base analysis”, Journal of Marketing Research, Vol.42(4),pp. 415-430 , 2005.

13. Ge, Y., Xiong, H., Tuzhilin, A., Xiao, K., Gruteser, M., & Pazzani, M., “An energy-efficient mobile recommender system”, 16th ACM SIGKDD international conference on Knowledge discovery and data mining , pp. 899-908, 2010.

14. Goldberg, D., Nichols, D., Oki, B. M., & Terry, D. “Using collaborative filtering to weave an information tapestry”, Communications of the ACM, Vol. 35(12), pp. 61-70 , 1992.

15. Gomez-Uribe, C. A., & Hunt, N.,“The netflix recommender system: Algorithms,business value, and innovation”, ACM Transactions on Management Information Systems (TMIS) , Vol.6(4) , pp. 1-19 , 2015.

16. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S., “ Neural Collaborative Filtering”, 26th International conference on World Wide Web , pp. 173-182 , 2017.

17. Jannach, D., Zanker, M., & Friedrich, G., “Recommender Systems”, Cambridge University Press , England , 2010.

18. Kent, A., Berry, M. M., Luehrs Jr, F. U., & Perry, J. W.,“Machine literature searching VIII. Operational criteria for designing information retrieval systems”, American documentation, Vol.6(2) , pp. 93-101 , 1955.

19. Kohavi, R., & Longbotham, R., “Online Controlled Experiments and A/B Testing”, Vol.7(8) , 2017.

20. Lee, S., Yang, J. H., & Park, S. Y.,“Discovery of hidden similarity on collaborative filtering to overcome sparsity problem”, International Conference on Discovery Science , pp. 396-402, 2004.

21. Levandowsky, M., & Winter, D., “Distance between sets”, Nature , Vol.234(5323) , pp. 34-35 , 1971.

22. Levens, M. P.,“Marketing: Defined, explained, applied”, Pearson College Div ,New Jersey, USA , 2012.

23. Linden, G., Smith, B., & York, J., Amazon.com recommendation - Item-to-item collaborative filtering. Ieee Internet Computing, 7(1), pp. 76-80, 2003.

24. MacQueen, J.,“Some methods for classification and analysis of multivariate observations”, 5th Berkeley symposium on mathematical statistics and probability, pp. 281-297, 1967.

25. Madhavaram, S., & Hunt, S. D., “The service-dominant logic and a hierarchy of operant resources: developing masterful operant resources and implications for marketing strategy”, Journal of the academy of marketing science, Vol. 36(1), pp. 67-82, 2008.

26. Mitchell, T. M.,“Machine learning”, McGraw-Hill, Inc., NY, United States ,1997.

27. Oshri, G.,“RFM: A Simple And Powerful Approach To Event Modeling”, Retrieved from http://ilikesqldata.com/rfm-a-simple-and-powerful-approach-toevent-modeling/, 2016.

28. Paul Resnick, & Varian, H. R.,“Recommender systems”, Communications of the ACM , pp. 56-58, 1997.

29. Powers, D. M.,“Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation”, Journal of Machine Learning Technologies ,Vol.2(1), pp. 37-63 , 2011.

30. Pride, W. M., O. C. Ferrell, B. L., Schembri, S., Niininen, O., & Casidy, R., “Marketing Principles, 3rd Asia-Pacific Edition.”, Cengage Learning AUS, Australia , 2018.

31. R. Agrawal, R. S., “Fast Algorithms for Mining Association Rules”, 20th International Conference on very large Data Bases, Santiago, Chile, pp. 407–419, 1994.

32. Reinschmidt, J., Gottschalk, H., Kim, H., & Zwietering, D., “Intelligent miner for data: enhance your business intelligence”, IBM international technical support organization (IBM Redbooks), NY, United States., 1999.

33. Resnick, P., lacovou, N., Suchak, M., & Bergstrom, P., & Riedl, J., “GroupLensan open architecture for collaborative filtering of netnews”, 1994 ACM conference on Computer supported cooperative work, pp. 175-186, 1994.

34. Rousseeuw, P. J., “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis”, Journal of Computational and Applied Mathematics, 20, pp. 53-65 , 1987.

35. Rumelhart, D. E., Hinton, G. E., & Williams, R. J.,“Learning representations by back-propagating errors”, Nature , 323(6088), pp. 533-536, 1986.

36. Salton, G. M., & McGill, M. J., “Introduction to Modern Information Retrieval”, NY,United States: McGraw-Hill, Inc., 1986.

37. Sarwar, B., Karypis, G., Konstan, J., & Riedl, J., “Application of dimensionality reduction in recommender system-a case study”, Minnesota Univ Minneapolis
Dept of Computer Science , 2000.

38. Shannon, C. E.,“A mathematical theory of communication”, Bell system technical journal, 27(3) , pp. 379-423 , 1948.

39. Sarwar, B., Karypis G., Konstan, J., and Ried, J.,“Analysis of recommendation algorithms for e-commerce”, Proceedings of the Second ACM Conference on Electronic Commerce, Minneapolis, pp. 158-167, 2000.

40. Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., . . . Hassabis, D., “Mastering the game of Go with deep neural networks and tree search”, Nature, Vol.529(7587), pp. 484-489, 2016.

41. Stone, M., “Cross-Validatory Choice and Assessment of Statistical Predictions”, Journal of the Royal Statistical Society .Series B (Methodological), Vol.36(2) , pp. 111-133 , 1974.

42. Tan, P.-N., Steinbach, M., & Kumar, V.,“Introduction to Data Mining ”, Addison-Wesley , United States , 2005.

43. Van Der Malsburg, C.,“Frank Rosenblatt : Principles of neurodynamics: Perceptrons and the theory of brain mechanisms ”, Springer , Berlin, Heidelberg , pp. 245-248 ,1961.

44. Vitale, R. P., & Giglierano, J. J., “Business to business marketing: analysis & practice in a dynamic environment”, South-Western/Thomas Learning ,Florida ,United States , 2002.