近年來，資料採礦引起了資訊產業界的極大關注，其主要原因是存在大量資料，可以廣泛使用，並且迫切需要將這些資料轉換成有用的資訊和知識。它利用了來自如下一些領域的思想：(1) 來自統計學的抽樣、估計和假設檢驗，(2)人工智慧、模式識別和機器學習的智慧演算法、建模技術和學習理論。資料採擷也迅速地接納了來自其他領域的思想，這些領域包括最優化、進化計算、資訊理論、信號處理、視覺化和資訊檢索。一些其他領域也產生重要的協助作用。尤其是，需要資料庫系統提供有效的存儲、索引和查詢處理支援。源於高性能（並行）計算的技術在處理海量資料集方面常常是重要的。分散式技術也能幫助處理海量數據，並且當資料不能集中到一起處理時更是至關重要。
本論文以四篇研究，實證資料採礦技術對人們在各種領域的貢獻，期望能幫助政府相關單位或企業決策者提升管理效能。
本論文第一篇文章應用三種不同的果蠅演算法(Fruit Fly Optimization Algorithm，FOA)應用於物流選址最佳化問題，我們應用原始的果蠅演算法、修正型果蠅演算法及混沌果蠅演算法針對物流中心選址的收益模型與成本模型進行尋優運算，以確定最佳的物流配送中心的位置，節省企業在物流配送工作的成本。研究結果發現，CFOA演算法的運算時間相對於FOA與MFOA較長，但反覆運算搜尋配送中心地址所獲得的收益相對於FOA與MFOA較高且所花費的成本相對於FOA與MFOA較低，因此建議未來企業主及研究人員可採用CFOA反覆運算搜尋配送中心地址。
第二篇文章為應用分量回歸探討企業採取精實服務之結果是否能提升整體服務之績效。本研究首先進行灰關聯分析，以此了解採取三個不同階段的餐飲服務流程，服務績效是否有差異；也就是採取精實服務流程，是否能提升績效。然後，我們再以分量回歸探討影響精實服務績效的主因。最後，我們將灰關聯分析之績效值，以二分類方式區分為績效佳與不佳兩類，再以FOAGRNN、AFSAGRNN、PSOGRNN以及GRNN，建立精實服務績效預警模型，並且與分量回歸模型作比較。由分析結果顯示，經由最佳化餐點製作與服務的產出，的確能提高整體生產與服務之績效。
第三篇係應用分量回歸進行銀行服務滿意度分析，以灰關聯分析台灣地區公民營銀行服務滿意度之調查問卷數據，以瞭解近幾年來滿意度績效排名前三名及最後三名之銀行；然後再採用分量迴歸探討哪些問項在何種分量下會顯著影響銀行服務滿意度績效，最後本論文以7種分量迴歸，建立銀行服務滿意度績效偵測模型。由分量迴歸分析發現，服務速度快速、調整利率能照顧客戶、業務錯誤發生次數及服務據點便利，不論是高低分量均會影響銀行服務滿意度績效；而7種分量回歸模型中，以Q50分量回歸模型的偵測能力最好。
第四篇係應用社會網絡分析軟體UCINET分析國內旅遊業組織成員間網絡互動影響知識資源。本論文以國內知名的易遊網行銷客服部門內30位成員為研究對象，分別進行網絡中心度分析法、網絡派系分析法以及MRQAP法進行分析。分析結果顯示, 該公司行銷服務部門成員在諮詢網絡、信任網絡及認知網絡的對偶互動程度，會影響知識資源(Know How)的交流程度，即各成員間在組織網絡互動愈密切，對於知識資源的交換便愈頻繁。
依上述四篇研究結果均顯示，以各種資料採礦技術可以輕易地分析各種領域及各種類型的問題，並且皆能得到良好的分析結果。由於各種資料採礦技術，各有不同的特性與用途，因此研究人員與相關學者必須加以深入研究，遇到各種問題能輕易判斷出該用何種資料採礦技術，方可針對各種不同問題提出解決方案，並獲得良好的分析成果。

Recently, data mining has attracted huge attention from whole information industry. The reason is that there are many data which can be widely used and we need to transform these data into useful information and knowledge urgently. Data mining takes advantage of the ideas from the following areas: (1) sampling of statistics, estimation and hypothesis testing, (2) the wisdom of the artificial intelligence, pattern recognition, AI algorithms, modeling technique and learning theory. Data mining also received ideas from other areas quickly, including optimization, evolutionary computation, information theory, signal processing, visualization and information retrieval. Some other areas also gave assistance to the development of data mining. To be more specific, effective data storage, database systems and query processing support are essential for data mining. The technology which derives from high-performance (parallel) computing has played a vital role in the process of massive data set. Distributed technique also contributed to process big dataset, especially when data set is distributed among more than one computer system.
This dissertation presented the following four studies aiming to testify the contributions of data mining technology in various fields of study. It is expected that this dissertation can provide the government, related units or enterprises decision makers with some useful advices in improving their management efficiency.
The first article adopted three different variants of Fruit Fly Algorithm (FOA) to solve the logistics location optimization problem. We applied the original Fruit Fly Algorithm, Modified Fruit Fly Algorithm and Chaotic Fruit Fly Algorithm (CFOA) to find the optimal logistics distribution center which took the income model and cost model of the location of a logistics center into account. The results revealed that the computation time of CFOA is longer those of the FOA and MFOA. However, the benefit gained through repeated searching for the location of a logistics distribution center was higher than those of the FOA and MFOA. Thus it is recommended that entrepreneurs and researchers use CFOA to find the optimal location of a distribution center in the future.
The second article discusses whether the adoption of the lean service could yield an improved service performance of an enterprise. First, grey relational analysis was conducted to examine whether there exists any difference in service performance of the three stages of a catering service process. In other words, this study explores whether the adoption of a lean service process would contribute to the improvement of the service performance of the catering service. Next, the main factors affecting the lean service performance was identified through quantile regression. Finally, performances were categorized as good or poor according to their performance values from the grey relational analysis. The results showed that the lean service has improved the performance of the catering production and service process of the catering service. Finally, early warning models for lean service performance were built with FOAGRNN, AFSAGRNN, PSOGRNN and GRNN, which were compared with the quantile regression model.
The third article studied the service satisfaction of banks through quantile regression. Grey relational analysis was employed to analyze the data of the questionnaires for the service satisfaction of public and private banks in Taiwan. Next, quantile regression was used to identify the items that affected the service satisfaction of the banks within each quantile. Finally, the detection models of the bank service satisfaction were built with seven quantile regression models. According to quantile regression, items of service speed, interest rate adjustment with clients taken into consideration, service error frequency and service point convenience affect the service satisfaction of banks for all high and low quantiles of service satisfaction. Besides, of the seven quantile regression models, the Q50 has the highest prediction accuracy.
The fourth article analyzed the impact of the Internet interaction among the members of a domestic tourist organization using UCINET. With the 30 members of the marketing service department of the well-known company EzTravel in Taiwan as the subjects of study, this study conducted analysis through network centrality, network fraction respectively. The results showed that the degree of dual interaction among the members of the marketing service department of the company in consultation network, trust network and cognitive network affected the interaction degree of knowledge resources. In other words, the more closely the members interacted with each other in the organizational network, the better the knowledge resources were exchanged.
The results of the above four studies show that data mining technologies can be easily used in analyzing problems of different fields with satisfactory results. With different characteristics and functions of different data mining technologies, researchers and scholars must conduct in-depth studies so that they can decide which data mining technique to use to deal with a problem in hand. Only in this way can they propose appropriate solutions to various problems and get satisfactory results.

[1] Deng, J. (1982). The control problems of grey system. System & Control Letters, 1(5), 288–294.
[2] Koenker, R. and Bassett, G. (1978). Regression quantiles. Econometrica, 46, 33-50.
[3] Davino, C., Romano, R. and Næs, T. (2015). The use of quantile regression in consumer studies. Food Quality & Preference, 40(40), 230-239.
[4] Marjanen, H., Engblom, J. and Malmari, M. (2013). Viability of demographic and behavioural independent variables in quantile regression models in predicting retail patronage. International Review of Retail Distribution & Consumer Research, 23(5), 523-536.
[5] Santos, C. and Cabral Vieira, J. (2012). An analysis of visitors' expenditures in a tourist destination: OLS, quantile regression and instrumental variable estimators. Tourism Economics, 18(3), 555-576.
[6] Specht, D.F. (1990). Probabilistic neural networks and the polynomial adaline as complementary techniques for classification. IEEE Trans. on Neural Networks, 1(1), 111–121.
[7] Specht, D.F. (1991). A general regression neural network. IEEE Tras. Neural Networks, 2(6), 568–576.
[8] Yeh, Y.C. (1998). The application of neural network. Julin Book Co., Ltd.
[9] Li, H.Z., Guo, S., Li, C.J. and Sun, J.Q. (2013). A hybrid annual power load forecasting model based on generalized regression neural network with fruit fly optimization algorithm. Knowledge-Based Systems, 37(2), 378-387.
[10] Jiang, P., Zeng, Z., Chen, J. and Huang, T. (2014). Generalized regression neural networks with k -fold cross-validation for displacement of landslide forecasting. Lecture Notes in Computer Science, 8866, 533-541.
[11] Chandrasekaran, R. and Manimannan, G. (2013). Use of generalized regression neural network in authorship. Clinical Orthopaedic Surgery, 62(4), 7-10.
[12] Eberhart, R. and Kennedy, J. (1995). A new optimizer using particle swarm theory. Paper presented at the Micro Machine and Human Science, 1995. MHS ‘95., Proceedings of the Sixth International Symposium on.
[13] Li, C.Q., Xu, S.P., Li, W. and Hu, L.G. (2012). A novel modified fly optimization algorithm for designing the self-tuning proportional integral derivative controller. Journal of Convergence Information Technology, 7(16), 69-77.
[14] Yuan, X.F., Liua, Y.M., Xianga, Y.Z. and Yan, X.G. (2015). Parameter identification of BIPT system using chaotic-enhanced fruit fly optimization algorithm. Applied Mathematics and Computation, 268, 1267–1281.
[15] Li, P., Zhang, T.T., Liang, T.G. and Zhang, K.H. (2015). Missing value imputation method based on density clustering and grey relational analysis. International Journal of Multimedia and Ubiquitous Engineering, 10(11), 133-142.
[16] Li, W.W., Li, Y.Q. and Tang, C.B. (2016). Web service selection method based on grey relational analysis. International Journal of Hybrid Information Technology, 9(3), 117-124.
[17] Guoping, Z. Longlong, Q. and Ningbo, W. (2012). Technology of qos evaluation based grey system theory. 2nd International Conference on Computer Science and Network Technology, 2012 December 29-31; Changchun, China.
[18] Xu, X.H., Pan, Z.J., Xi, Y.Q. and Chen, L. (2012). Incremental particle swarm optimization. Physics Procedia, 24, Part B, 1369-1376.
[19] Xue, Z., Du, P. and Su, H. (2014). Harmonic analysis for hyperspectral image classification integrated with pso optimized svm. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 7(6), 2131-2146.
[20] Gong, X., Liu, P. and Jian, J. (2015). Based on the integrated neutral-network optimized by PSO multiple harmful gas detection system. Chinese Journal of Sensors & Actuators, 28(6), 938-942.
[21] Li, X.L., Shao, Z.J. and Qian, J.X. (2002). An optimization searching model based on animal autonomous body: artificial fish swarm algorithm. System Engineering Theory and Practice, 22(11), 32–38.
[22] Pan, W.T. (2012). A new fruit fly optimization algorithm: Taking the financial distress model as an example. Knowledge-Based Systems, 26, 69-74.
[23] Womack, J.P. and Jones, D.T. (2006). Lean thinking. New York, United States: Simon & Schuster.
[24] Wang, G.X. (2013). Analysis on power pathway measurement in public policy: the case study of national health insurance’s crucial policy events. Public Administration & Policy, (57), 37-90.
[25] Wang, G.X. and Hsung, R.M. (2012). Using network analysis for researching brokerage roles in policy processes: the case of Taichung City’s development domain before and after the lifting of martial law. Conference of Taiwan Association for Schools of Public Administraion & Affairs.
[26] Chen, J.T. (2004). The emergences and consequences of intra-organizational social networks--social capital perspective. Doctoral dissertation. Graduate Institute of Human Resource Management, National Sun Yat-sen University.
[27] Hung, S.Y., Huang, Y.W. and Lai, H.M. (2015). A meta-analysis of critical factors for knowledge sharing. Journal of Information Management, 22(4), 403-444.
[28] Nonaka, I., Toyama, R. and Konno, N. (2000). SECI,Ba,and Leadership:a unified model of dynamic knowledge creation. Long Range Planning, 33, 5-34.
[29] Yu, Y.N. and Yuan, M.S. (2013). A network analysis of the teachers and graduate students’research topics in the field of mass communication. Journal of InfoLib & Archives, 11(1), 131-165.
[30] Thomas, C.W. (1998). Maintaining and restoring public trust in government agencies and their employees. Administration and Society, 30, 166-193.
[31] Burt, R.S. (1992). Structural holes: the social structure of sompetition. Harvard University Press: Cambridge, MA.
[32] Krackhardt, D. (1993). MRQAP: Analytic versus permutatioin solutions. Working paper, Carnegie Mellon University.
[33] Nonaka, I. and Takeuchi, H. (1995). The knowledge-creating company. NY: Oxford University Press, Inc.
[34] Weng, C.S. (2011). The exploration and exploitation of technologies in between of the core and the periphery－insurance business method patents. Journal of Management, 28(3), 219-246.
[35] Krackhardt, D. (1987). QAP Partialling as a Test of Spuriousness. Social Networks, 9, 171-186.
[36] Kurvers, R.H.J.M., Adamczyk, V.M.A.P., Kraus, R.H.S., Hoffman, J.I., Van Wieren, S.E., Van der Jeugd, H.P., Amos, W., Prins, H.H.T. and Jonker, R.M. (2013). Contrasting context dependence of familiarity and kinship in animal social networks. Animal Behaviour, 86, 993-1001.
[37] Gao, F. (2014). Matlab intelligent algorithm super learning manual. Post & Telecom Press.
[38] Liu, Q., Odaka, T., Kuroiwa, J., Shirai, H. and Ogura, H. (2014). A new artificial fish swarm algorithm for the multiple knapsack problem. Ieice Transactions on Information & Systems, E97.D(3), 455-468.
[39] Neshat, M., Sepidnam, G., Sargolzaei, M. and Toosi, A.N. (2014). Artificial fish swarm algorithm: a survey of the state-of-the-art, hybridization, combinatorial and indicative applications. Artificial Intelligence Review, 42(4), 965–997.
[40] Fang, R.X., Yao, W.H. and Zhang, X.D. (2015). Short-term traffic flow forecasting based on svr with improved artificial fish swarm algorithm. Applied Mechanics & Materials, 734(8), 508-514.
[41] Zhang, Y., Cui, G., Wu, J., Pan,W.T. and He, Q. (2016). A novel multi-scale cooperative mutation Fruit Fly Optimization Algorithm. Knowledge-Based Systems, 114, 24-35.
[42] Li, J.Q., Pan, Q.K. and Mao, K. (2016). A hybrid fruit fly optimization algorithm for the realistic hybrid flowshop rescheduling problem in steelmaking systems. IEEE Transactions on Automation Science & Engineering, 13(2), 932-949.
[43] Zhao, J. and Yuan, X. (2016). Multi-objective optimization of stand-alone hybrid PV-wind-diesel-battery system using improved fruit fly optimization algorithm. Soft Computing, 20(7), 2841-2853.
[44] Ding, S., Zhang, X. and Yu, J. (2016). Twin support vector machines based on fruit fly optimization algorithm. International Journal of Machine Learning and Cybernetics, 7(2), 193-203.
[45] Wang, W. and Liu, X. (2015). Melt index prediction by least squares support vector machines with an adaptive mutation fruit fly optimization algorithm. Chemometrics & Intelligent Laboratory Systems, 141:79-87.
[46] Chen, Z.X. (2010). Fundamentals of production and operation management. Electronic Industry Press.
[47] Gerrard, P. and Cunningham, B. (2001). Bank service quality: a comparison between a publicly quoted bank and a government bank in Singapore. Journal of Financial Services Marketing, 6(1), 50–66.
[48] Ennew, C.T. and Binks, M. R. (1996). Theimpact of service quality and service characteristics on customer retention: small businesses and their banks in the UK. British Journal of Management, 7(3), 219–230.
[49] Tosun, N. (2006). Determination of optimum parameters for multiperformance characteristics in drilling by using grey relational analysis. International Journal of Advanced Manufacturing Technology, 28(5-6), 450–455.
[50] Nagpal, G., Uddin, M. and Kaur, A. (2012). A hybrid technique using grey relational analysis and regression for software effort estimation using feature selection. International Journal of Soft Computing and Engineering, 1(6), 20–27.
[51] Hasani, H., Tabatabaei, S. A. and Amiri, G. (2012). Grey relational analysis to determine the optimum process parameters for open-end spinning yarns. Journal of Engineered Fibers and Fabrics, 7(2), 81–86.
[52] Pan, W.T. (2014). Mixed modified fruit fly optimization algorithm with general regression neural network to build oil and gold prices forecasting model. Kybernetes, 43(7), 1053–1063.
[53] Borgatti, S. P., Jones, C. and Everett, M. G. (1998). Network measures of social capital. Connections, 21(2), 27–36.
[54] Nahapiet, J. and Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. The Academy of Management Review, 23(2), 242 - 266.
[55] Bradley, A.P. (1997). The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognition, 30, 1145–1159.