隨著資訊科技(Information Technology)的快速進步且操作性資訊系統的普遍，企業資料庫迅速累積了龐大的資料量。如果能適時針對這些資料做分析，將資料轉換成有用資訊，將成為企業競爭的利器。而最常被用來分析企業資料的技術即為資料探勘(Data Mining)，但一般傳統資料探勘技術大多著重於商品銷售量、顧客喜好商品類別及服務之探勘，往往忽略了企業獲利的部份。而企業經營最終目的還是在於獲利，因此本研究以企業獲利為觀點，並且為了解決數值型資料易落入模糊區間範圍的問題，將利用模糊資料探勘(Fuzzy Data Mining)技術應用在旅遊業與賣場業來探勘出能提高企業利潤的消費組合，以期提供企業提升己身獲利能力的參考依據。
在過去，對於資料探勘的相關研究大多以二維屬性的交易資料為多數。然而，經由探勘步驟所得到的關聯法則(Association Rule)之存在通常與當時的環境有高度的關聯性，因此本研究在旅遊業應用例子中將使用多重維度資料，如遊客所停留的景點區域與當時到達景點位置的時段與其停留時間來探勘，以期在實務應用上能更符合當時環境之實際情況。
在賣場業應用例子的高銷售率且高獲利率商品組合探勘部份，主要是利用正規化方式來求得成本與利潤相對於全體所佔的比重與其背後所代表的數值意義，例如可瞭解在全體商品中，哪個商品項目是具有潛在的高利潤，同時避免游離數值的問題產生；在低銷售率且高獲利率商品組合探勘部份，本研究利用基因演算法(Genetic Algorithm, GAs)來不斷地加以訓練歸屬函數(Membership Function)之參數值，藉由演化(Evolution)過程來產生族群子代(Population)並求得最佳可行解，以達到調整出近似最佳的歸屬函數之區間形狀，解決以往人們由經驗累積所形成的主觀意識去判斷歸屬函數的形狀其可能所會帶來探勘結果的錯誤。

A huge amount of data has been accumulated in the enterprise databases due to the wide use of information systems today. An enterprise may have more competitive advantages if the useful information can be discovered from the collected data. Data mining thus is widely applied for the purpose. Previous studies, nevertheless, mine either the occurrence-associations of the products or the quantities of the associated items without considering the genuine concern of the enterprise – the profit. In this thesis, the numeric data that is central to the profit is transformed into semantically meaningful terms, using fuzzy techniques. Fuzzy data mining algorithms are proposed to solve the profit mining problems in both tourism and supermarket applications.
Making recommendations for customers as added services may increase the competitive advantages of a tourism operator. Most association-rule mining algorithms are limited to one or two dimensional transactional data so that the results for recommendations might be unsatisfactory. A high quality tour recommendation should be generated by considering the places, times, durations, and sequences of tourism spots altogether. Therefore, a multi-dimensional sequential pattern mining algorithm with fuzzy semantics is proposed to discover effective tour recommendations in this study.
The profit mining in supermarket applications is emphasized on the high-profit items. For items of high-profit rate and high-sale rate, a normalization technique is applied to extract the true weight and meaning of the item in proportion to the whole transaction in the aspects of costs and profits. The item-sets that generate the majority of profits for the enterprise and the outliers can be identified accordingly without bias. For items of high-profit rate but low-sale rate, a generic algorithm is applied to eliminate the subjective determination from human experiences. The parameters of the membership functions are trained using the genetic algorithm to generate the initial populations that evolve into the optimal solution for profitable items. The shapes are also adjusted by the generic algorithm on the basis of gross profits. Human errors can be minimized and the high profit item-sets, either high or low sale rate, can be discovered.

摘要 I
Abstract III
誌謝 V
目錄 VII
圖目錄 X
表目錄 XII
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 5
1.3 研究範圍 5
1.4 研究步驟 7
1.5 論文架構 10
第二章 文獻探討 12
2.1 資料探勘 12
2.1.1 關聯法則 13
2.1.2 序列型樣 14
2.2 模糊理論 15
2.2.1 模糊數 15
2.2.2 歸屬函數與語意變數 17
2.3 模糊資料探勘 18
2.4 模糊C均值群聚 22
2.5 基因演算法 26
2.5.1 演化流程 26
2.5.2 不可行解區域處理策略 28
第三章 模糊資料探勘之應用-以旅遊業為例 31
3.1 探勘流程 32
3.1.1 利用模糊C均值群聚技術來劃分區域 36
3.1.2 使用模糊資料探勘技術來探勘旅遊序列型樣 37
3.2 應用例子說明 38
3.3實驗結果 43
3.4小結 46
第四章 模糊資料探勘應用-以高獲利率商品組合探勘為例 48
4.1 「高銷售率且高獲利率」商品組合之探勘 49
4.1.1 探勘流程 49
4.1.2 應用例子說明 53
4.1.3 實驗結果 58
4.2 「低銷售率且高獲利率」商品組合之探勘 59
4.2.1 探勘流程 59
4.2.2 應用例子說明 65
4.2.3 實驗結果 70
4.3 小結 75
第五章 結論 77
5.1 研究貢獻 77
5.2 未來研究方向 79
參考文獻 81
個人簡歷 92

[1]王建驊(2005)，使用模糊資料探勘評估線上遊戲消費者之偏好，碩士論文，玄奘大學資訊管理系所，新竹。
[2]克里斯•安德森(2006), 長尾理論：打破20/80法則的新經濟學，天下文化出版社。
[3]宋紹儀(2004)，以資料探勘之應用-設計員工知識管理資訊系統，碩士論文，國立台北大學統計學系研究所，台北。
[4]李權源(2005)，平行式工具機之基因法參數優化與模糊法扭力控制平行式工具機之基因法參數優化與模糊法扭力控制，碩士論文，國立交通大學機械工程系所，新竹。
[5]奇摩知識+(2006.12.10)，http://tw.search.yahoo.com/search/。
[6]胡宜中(2002)，模糊知識探勘之多屬性決策，碩士論文，國立交通大學資訊管理系所，新竹。
[7]范義明(2002)，Linux作業系統下異常偵測之研究，碩士論文，大葉大學資訊管理系所，彰化。
[8]陳永信(2004)，以資料探勘法鏈結顧客知識管理於創新性產品開發專案之研究，博士論文，國立交通大學工業工程與管理研究所，新竹。
[9]陳俊豪(2004)，利用遺傳演算法探勘隸屬函數與模糊關聯規則，碩士論文，義守大學資訊管理系所，高雄。
[10]游宮華(2005)，FEIMA:以模糊限制為基礎之有效互動序列型樣探勘研究，碩士論文，元智大學資訊工程系所，桃園。
[11]黃振益(2003)，資料挖掘於電信服務客戶流失之研究，碩士論文，中國文化大學資訊管理系所，台北。
[12]蕭廣佑(2003)，應用模糊資料探勘於網頁瀏覽記錄之研究，碩士論文，南台科技大學資訊管理系所，台南。
[13]蘇建源(2004)，模糊邏輯與資料探勘技術為基礎在顧客關係管理上之研究與應用，碩士論文，南華大學資訊管理研究所，嘉義。
[14]龔惠盈(2004)，資料採礦在產品庫存價值分析之研究－以A公司為例，碩士論文，元智大學工業工程與管理研究所，桃園。
[15]R. Agrawal and R. Srikant(1995), “Mining Sequential Patterns,” Proceedings of the 11th International Conference on Data Engineering, pp. 3-14.
[16]R. Agrawal, T. Imielinski, and A. N. Swami(1993), “Mining Association Rules between Sets of Items in Large Databases,” Proceedings of the ACM SIGMOD Conference on Management of Data, Washington, USA, Vol. 22, No. 2, pp. 207-216.
[17]D. Arotaritei(2002), “Mining Fuzzy Association Rules in Incomplete Database,” Proceedings of the IEEE International Conference on Fuzzy Systems, Vol. 1, pp. 267-271.
[18]M. J. A. Berry and G. Linoff(1997), Data Mining Techniques: For Marketing Sale and Customer Support, John Wilely and Sons, Inc., Canada.
[19]J. C. Bezdek(1973), Fuzzy Mathematics in Pattern Classification, Math Center, Cornell University Lthaca.
[20]J. C. Bezdek(1981), “Pattern Recognition With Fuzzy Objective Function Algorithm,” Advanced Applications in Pattern Recognition, New York: Plenum Press.
[21]S. G. Cao, N. W. Rees, and G. Feng(1996), “Quadratic Stability Analysis and Design of Continuous-Time Fuzzy Control Systems,” Soft Computing: Elements of Learning Systems, Vol. 27, No. 2, pp. 193-203.
[22]K. C. C. Chan and W. H. Au(1997), “Mining Fuzzy Association Rules,” Proceedings of the 6th International Conference on Information and Knowledge Management, pp. 209-215.
[23]R. L. P. Chang and T. Pavliddis(1997), “Fuzzy Decision Tree Algorithms,” Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, Vol. 7, No. 1, pp. 28-35.
[24]Y. Chang, J. L. Chen, and M. L. Tsay(1999), “Landscape Image Analysis Using Fuzzy Adaptive Resonance Theory,” Proceedings of the IEEE Conference on Systems, Man, and Cybernetics, Tokyo, Japan, Vol. 3, No. 21, pp. 84-89.
[25]C. H. Chen, T. P. Hong, and V. S. Tseng(2005), “Analyzing Time-Series Data by Fuzzy Data Mining Technique,” Proceedings of the IEEE International Conference on Granular Computing, Vol. 1, No. 24, pp. 112-117.
[26]G. Chen and Q. Wei(2002), “Fuzzy Association Rules and the Extended Mining Algorithms,” Information Sciences, Vol. 147, pp. 201-228.
[27]M. S. Chen, J. S. Park, and P. S. Yu(1998), “Efficient Data Mining for Path Traversal Patterns,” IEEE Transactions on Knowledge and Data Engineering, Vol. 10, No. 2, pp. 209-221.
[28]Q. Chen, H. Jianghong, W. He, K. M, and Y. Lai(2005), “Utilize Fuzzy Data Mining to Find the Travel Pattern of Browsers,” Proceedings of the 5th International Conference on Computer and Information Technology, pp. 228-232.
[29]R. S. Chen, R. C. Wu, and J. Y. Chen(2005), “Data Mining Application in Customer Relationship Management of Credit Card Business,” Proceedings of the 29th International Conference on Annual International Computer Software and Applications, Vol. 2, pp. 39-40.
[30]Y. L. Chen, S. S. Chen, and P. Y. Hsu(2002), “Mining Hybrid Sequential Patterns and Sequential Rules,” Information Systems, Vol. 27, No. 5, pp. 345-362.
[31]Y. Q. Chen and L. F. Hu(2005), “Study on Data Mining Application in CRM System Based on Insurance Trade,” Proceedings of the 7th International Conference on Electronic Commerce, New York, Vol. 113, pp. 839-841.
[32]W. Chimphlee, A. H. Abdullah, M. N. M. Sap, S. Chimphlee, and S. Srinoy(2005), “Integrating Genetic Algorithm and Fuzzy C-means for Anomaly Detection,” IEEE Indicon Conference, pp. 575-579.
[33]M. W. Craven and J. W. Shavlik(1997), “Using Neural Networks for Data Mining,” Future Generation Computer System, Vol. 13, pp. 221-229.
[34]D. Dubois and H. Parade(1980), Fuzzy Sets and Systems: Theory and Application, Academic Press Inc.
[35]S. Eschrich, J. Ke, L. O. Hall, and D. B. Goldgof(2001), “Fast Fuzzy Clustering of Infrared Images,” Proceedings of the 20th International Conference on North American Fuzzy Information Processing Society, Vol. 2, pp. 1145-1150.
[36]U. Fayyed, S. G. Piatetsky, and S. Padhraic(1996), “From Data Mining to Knowledge Discovery in Databases,” American Association for Artificial Intelligence, Vol. 17, No. 3, pp. 37-54.
[37]M. Gen and R. Cheng(2000), Genetic Algorithm and Engineering Optimization, John Wiley & Sons, New York.
[38]D. E. Goldberg(1989), Genetic Algorithms in Search, Optimization & Machine Learning, Addison Wesley.
[39]P. V. Gorsevskil, P. E. Gesslerl and P. Jankowski(2003), “Integrating a Fuzzy K-means Classification and a Bayesian Approach for Spatial Prediction of Landslide Hazard,” The Journal of Geograph System, Vol. 5, pp. 223-251.
[40]F. H. Grupe and M. M. Owrang(1995), “Database Mining Discovering New Knowledge and Cooperative Advantage,” Information Systems Management, Vol. 12, No. 4, pp. 26-31.
[41]J. Han, J. Pei, H. Pinto, Q. Chen, U. Dayal, and M. C. Hsu(2001), “PrefixSpan: Mining Sequential Patterns Efficiently by Prefix-Projected Pattern Growth,” Proceedings of the 17th International Conference on Data Engineering, Heidelberg, Germany, pp. 215-224.
[42]J. W. Han and M. Kanber(2001), Data Mining Concepts and Techniques, Morgan Kaufmann Publishers.
[43]J. H. Holland(1975), Adaptation in Natural and Artificial System, University of Michigan Press.
[44]T. P. Hong, C. S. Kuo, and S. C. Chi(1999), “A Fuzzy Data Mining Algorithm for Quantitative Values,” Proceedings of the 3rd International Conference on Knowledge-Based Intelligent Information Engineering Systems, pp. 480-443.
[45]T. P. Hong, C. S. Kuo, and S. C. Chi(1999), “Mining Fuzzy Sequential Patterns from Quantitative Data,” Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, Vol. 3, pp. 962-966.
[46]T. P. Hong, C. S. Kuo, S. C. Chi and S. L. Wang(2002), “A Fuzzy AprioriTid Mining Algorithm,” International Mathematical Journal, Vol. 2, No. 11, pp. 1089-1117.
[47]T. P. Hong, K. Y. Lin and S. L. Wang(2002), “Mining Linguistic Browsing Patterns In the World Wide Web," Soft Computing, Vol. 6, No. 5, pp. 329-336.
[48]T. P. Hong, K. Y. Lin, and B. C. Chien(2003), “Mining Fuzzy Multiple-Level Association Rules from Quantitative Data,” Intelligence, Vol. 18, No.1, pp. 79-90.
[49]T. P. Hong, M. J. Chiang, and S. L. Wang(2002), “Mining Weighted Browsing Patterns with Linguistic Minimum Supports,” Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, Vol. 4, pp. 635-639.
[50]T. P. Hong, Y. C. Lee, and T. C. Wang(2006), “Mining Fuzzy Multiple-level Association Rules under Multiple Minimum Supports," Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, Vol. 5, pp. 4112-4117.
[51]Y. C. Hu, and G. H. Tzeng(2003), “Elicitation of Classification Rules by Fuzzy Data Mining,” Engineering Applications of Artificial Intelligence, Vol. 16, No. 7-8, pp. 709-716.
[52]S. C. Hui and G. Jha(1999), “Data Mining for Customer Service Support,” Information Management, Vol. 38, No. 1, pp. 1-13.
[53]A. Kaufmann and M. M. Gupta(1991), Introduction to Fuzzy Arithmetic: Theory and Application, Van Nostrand Reinhold.
[54]R. Koch(1998), The 80/20 Principle: The Secret of Achieving More with Less, Doubleday, New York City.
[55]C. M. Kuok, A. Fu, and M. Wong(1998), “Mining Fuzzy Association Rules in Databases,” SIGMOD Record , Vol. 17, No.1, pp. 41-46.
[56]J. W. T. Lee(2000), “An Ordinal Framework for Data Mining of Fuzzy Rules,” Proceedings of the 9th IEEE International Conference on Fuzzy System, pp. 399-404.
[57]G. S. Liang(1999), “Theory and Methodology: Fuzzy MCDM based on Ideal and Anti-Ideal Concepts,” European Journal of Operational Research, Vol. 112, pp. 682-691.
[58]W. Y. Liu, Z. W. Chen, P. Bai, S. F. Fang, and Y. Shi(2005), “A Kind of Improved Method of Fuzzy Clustering,” Proceedings of the International Conference on Machine Learning and Cybernetics, Vol. 5, pp. 2646-2649.
[59]J. Luo and S. M. Bridges(2000), “Mining Fuzzy Association Rules and Fuzzy Frequency Episodes for Intrusion Detection,” International Journal of Intelligent Systems, Vol. 15, pp. 687-703.
[60]M. Mitchell(1996), An Introduction to Genetic Algorithms, MIT press.
[61]T. Morzy, M. Wojciechnowski, and M. Zakrzewicz(1999), “Pattern-Oriented Hierarchical Clustering,” Proceedings of the 3rd East European Conference on Advances in Databases and Information System, Maribor, Slovenia, LNCS 1691, Springer-Verlag.
[62]T. Morzy, M. Wojciechowski, and M. Zakrzewicz(2001), “Scalable Hierarchical Clustering Method for Sequences of Categorical Values,” Proceedings of the 5th Pacific-Asia Conference on Knowledge Discovery and Data Mining, Kowloon, Hong Kong, LNAI 2035, Springer-Verlag.
[63]W. Pedrycz, “Conditional Fuzzy C-means(1996),” Pattern Recognition Letters, Vol. 17, pp. 625-632.
[64]D. L. Pham, “Spatial Models for Fuzzy Clustering,” Computer Vision and Image Understanding, Vol. 84, pp. 285-297.
[65]T. D. Pham(2001), “Image Segmentation using Probabilistic Fuzzy C-means Clustering Algorithm,” Proceedings of the International Conference on Image Processing, Vol. 1, pp. 722-725.
[66]D. Pyle(1991), Data Preparation for Data Mining, Morgan Kaufmann Publishers.
[67]E. H. Ruspini (1969), “A New Approach to Clustering,” Information Control, Vol. 15, pp. 22-32.
[68]R. Srikant, Q. Vu, and R. Agrawal(1997), “Mining Association Rules with Item Constraints,” Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining, California, pp. 67-73.
[69]P. S. M. Tsai and C. M. Chen(2004), “Mining Interesting Association Rules from Customer Databases and Transaction Databases,” Information Systems, Vol. 29, pp. 685-696.
[70]J. Wu and J. Li(2004), “Infrared Image Segmentation via Fast Fuzzy C-means with Spatial Information,” Proceedings of the IEEE International Conference on Robotics and Biomimetics, pp. 742-745.
[71]K. L. Wu and M. S. Yang(2002), “Alternative C-means Clustering Algorithm,” Pattern Recognition, Vol. 35, pp. 2267-2278.
[72]R. Wu, W. Tang, R. Zhao(2004), “An Efficient Algorithm for Fuzzy Web-Mining,” Proceedings of the IEEE International Conference on Information Reuse and Integration, pp. 576-581.
[73]M. S. Yang(1993), “On a Class of Fuzzy Classification Maximum Likelihood Procedures,” Fuzzy Sets and Systems, Vol. 58, pp. 365-375.
[74]L. A. Zadeh(1965), “Fuzzy sets,” Information and Control, Vol. 8, pp. 338-353.
[75]M. J. Zaki(2001), “SPADE: An Efficient Algorithm for Mining Frequent Sequences,” Proceedings of Machine Learning Journal, Special Issue on Unsupervised Learning, Vol. 42, pp. 31-60.