在電子商務環境中，各種電子銀行通路可藉由網際網路及網頁技術與客戶直接溝通，電子支付通路因而更加暢通。在此衝擊下，行庫及企業對扮演跨組織系統中帳務性資料交換及清算角色的金融電子資料交換通路該何去何從？是本研究探討的重點。本研究由三個方向來探討「金融EDI通路發展的期待」，從六個考量因素來探討「網際網路衝擊的考量」。研究顯示「金融EDI通路發展的期待」與「網際網路衝擊的考量」卡方檢定列聯相關結果顯示：三個期待方向皆呈統計顯著性。「維持現狀」之期待與「行庫及企業本身的特性」、「發展網際網路應用的成本考量」、「發展網際網路應用的速度考量」具相關性。「結合方式」之期待與「發展網際網路應用的成本考量」具相關性。「完全取代」之期待與「發展網際網路應用的速度考量」具相關性。研究亦顯示在「發展網際網路應用的成本考量」上，若有妥善解決之道，則金融EDI用戶會傾向透過開放的網際網路資訊高速公路來連上專屬的金融EDI通路來作業，否則金融EDI用戶會傾向仍維持採用原來的專屬金融EDI通路。在「發展網際網路應用的速度考量」上若有妥善解決之道，則金融EDI用戶會傾向把在專屬網路上的EDI及金融EDI作業方式完全轉移到開放的網際網路上進行。否則金融EDI用戶會傾向仍維持採用原來的專屈金融EDI通路。本研究附帶探討在「行業別」對「網際網路衝擊的考量」各考量變異數分析及T檢定發現：金融機構比非金融機構重視發展網際網路應用的優勢考量、速度考量、及安全考量。而非金融機構則比金融機構重視發展網際網路應用的阻力考量及成本考量。而各項衝擊考量中有明顯差異之子項則可提供作為兩類機構在發展網際網路相關應用之考量重點。在「職稱別」對「網際網路衝擊的考量」各考量變異數分析及T檢定發現：資訊單位對網際網路衝擊的認知普遍高於業務單位，而各項衝擊考量中有明顯差異之子項則可提供作為雙方在進行網際網路相關之資訊化業務溝通的討論重點。本研究結果可提供金融EDI政策制定單位、金融EDI服務提供單位對金融EDI定位及後續功能發展方向之參考，也可提供使用金融EDI的行庫及企業在推動金融EDI作業時，對相關的考量因素能有較明確的認知。

在統計學上，我們常使用皮爾森相關係數(Pearson’s Correlation Coefficient)來表達兩變數間線性關係的強度，同時也表達出關係之方向。傳統之相關係數所處理的資料都是明確的實數值，但是當資料是模糊數時，並不適合使用傳統的方法來計算模糊相關係數。而本研究探討區間模糊樣本資料值求得模糊相關係數，首先將區間型模糊資料分為離散型和連續型，提出區間模糊相關係數定義，並提出廣義誤差公式，將相關係數作合理的調整，使所求的出相關係數更加精確。在第三章我們以影響數學成就評量的因素，作實證研究分析，得出合理的分析。而此相關係數定義和廣義誤差公式也能應用在兩資料值為實數或其中一筆資料值為實數的情況，可以解釋更多在實務上所發生的相關現象。 / In the statistic research, we usually express the magnitude of linear relation between two variables by means of Pearson’s Correlation Coefficient, which is also used to convey the direction of such relation. Traditionally, correlation coefficient deals with data which consist of specific real numbers. But when the data are composed of fuzzy numbers, it is not feasible to use this traditional approach to figure out the fuzzy correlation coefficient. The present study investigates the fuzzy samples of interval data to find out the fuzzy correlation coefficient. First, we categorize the fuzzy interval data into two types: discrete and continuous. Second, we define fuzzy correlation with interval data and propose broad formulas of error in order to adjust the coefficient more reasonably and deal with it more accurately. In Chapter Three, we conduct empirical research by the factor which affects the evaluation of mathematical achievement to acquire reasonable analysis. By doing so, broad definition of coefficient and formulas of error can also be applied to the conditions of either both values of the data are real number or one value of the data is real number, and can explain more related practical phenomenon.

近年來由於政府實施無紙化及金融業者推行電子帳單的成效卓越，使得國內郵件的收寄量逐年下滑，郵局如何與民營業者競爭國內物流市場並達成盈餘目標，成為營運中不可忽視之因素。 傳統的郵局設點多依據公司規定與配合政府政策需求，甚少採用涉及複雜因素之區位分析進行選址。因此，如何有效且公正地評選郵局新設據點以提高收益，成為亟待解決之問題。 本研究目的在於提供高收益之郵局設點建議，我們提出一種評估中華郵政公司設點效益的方法，以國內郵局實際設點位置與相關空間資料來建置實驗模型。研究結果顯示，以本研究方法建立之預測模型可成功的提供中華郵政公司建議於何處新增據點可收最大功效。 我們首先蒐集中華郵政公司設點之鄰近區域資料，如競爭者設點數、人口因素、重要交通路口、郵件收寄量等。其次導入資料探勘技術分析影響郵件收寄量之因素，建立中華郵政公司設點收寄量預測模型。然後依照建立預測模型時所得到之區辨力分數，判斷採用何種資料探勘技術建立預測模型較適當。最後將所選定的預測模型套用於台北縣市各村里建物重心，透過環域資料分析以計算預估之收寄量，再整合各資料探勘技術之預測結果後推論出最佳設點建議。 實作中，以台北縣市資料來測試我們的方法。實驗數據顯示，我們的方法成功地找出十一個建議設點的村里，可提供給中華郵政公司作為高收益的設點建議。 / The amount of postal mail declines in recent years due to the efforts of paper-reduce policies implemented by the government, the industries, and the general publics. It becomes one of the important issues of the Chunghwa Post Company, to compete with other companies in domestic freight and mail services and to achieve the desired profits. Traditionally, the location of post offices were decided according to the government policies as well as the company regulations. The issues involved in the site selection analysis were seldom considered. Hence, developing an effective and fair mechanism to find the new post office locations that could improve the company’s surplus becomes an important problem to be solved. The purpose of this thesis is to provide recommendations to the post office site selection which will yield high profit to the company. We proposed a method to evaluate the effective profits that could be produced by a particular post office through the data mining techniques and the related GIS information. We first collect various data, such as neighborhood population, traffic flow, postal mail received at particular post office, competitor’s information, etc., and analyze these data using data mining techniques in order to establish prediction models. The most appropriate model was chosen to find the new post office sites. The Metropolitan Taipei area was chosen to illustrate our idea. The best sites for new post offices were selected through the buffering analysis as well as the data mining techniques. The experimental results show that our method can successfully find eleven locations which could generate most profit to Chunghwa Post Company if the new post offices were located in these places.

本文利用「華人家庭動態資料庫」RI1999與RI2003兩年資料，以多元羅吉特模型 (Multinomial Logit Model)，探討27歲至46歲的年輕家庭對於住宅權屬 (自有房屋、租屋、父母所有房屋) 的決策。 文中主要針對不同世代之重大生命歷程事件 (結婚生子) 對住宅權屬選擇的影響差異進行比較分析。研究結果發現，生命歷程事件 (結婚生子) 對自有房屋的影響力隨世代不同已逐漸減弱，但住在父母所有房屋的傾向卻愈來愈強烈。然而，小孩個數愈多，住在父母所有房屋機率增加、自有機率下降，婚後住宅權屬發生改變的機率也會增加。此外，研究亦發現，對新生代家庭來說，都會區家庭自有房屋的機率比非都會區家庭高，更凸顯出原生家庭的資源與支持對新生代家庭的重要性。

雲端運算和資料採礦已成為這二十一世紀的重要發展方向，綜觀現今各個生活層面，已漸漸的融合雲端計算的技術，故結合雲端運算已是一種趨勢。簡而談之，雲端運算是一種讓使用者更加地快速、便利又省成本的一種技術。而資料採礦方面，也已從先前的專門挖掘數字型態的資料，到現在多元的挖掘，像是文字、圖像採礦。資料採礦雖然比雲端運算發展的早，但是其功用是可以相輔相成的，有鑑於此，本研究係要發展出一資料採礦分析系統，使得使用者方便又簡易的操作。並針對特定的資料採礦分析方法-關聯規則及集群分析去研究，並利用Apriori 演算法及K-means方法，和Microsoft Excel VBA和R軟體共同結合出此資料採礦系統。

都市經濟發展同時卻也造成環境衝擊的影響，在永續發展理念下，經濟和環境議題調和為重要議題，過去評估都市永續性主要透過永續指標建構、生態足跡、能值分析等方法，但缺乏將環境和經濟兩者同時考量，因而有經濟-生態效率（Eco-efficiency）評估方式產生。本文嘗試利用此評估方式於探討台灣在民國85、90及95年經濟與環境變化情形，並分析哪些縣市為發展上具有效率以及影響縣市經濟-生態效率上差異性原因。本研究第一階段利用可處理多投入及多產出之資料包絡分析計算效率，第二階段則將以都市發展程度等為原因探討對於縣市效率值變動影響情形。 研究結果顯示，縣市中以北部縣市效率值較高而南部效率值較低，個別縣市觀察以台北市、新竹縣、新竹市、台中市、嘉義市和台東縣為三個年度中皆具有效率的縣市。而縣市欲提升效率值則需從汽柴油銷售量及用電量著手；此外，總用水量和二、三級產業就業員工數為縣市經濟-生態效率值之優勢因素。更近一步透過Tobit迴歸分析影響縣市效率值差異性原因，都市發展程度及污染性產業比例與效率值呈現負向變動，而每人每年可支配所得高低則呈正向變動。因此，建議政府對產業的汙染管制宜改善，而於都市發展程度較高的縣市提升能源使用效率並降低對環境影響，以提升縣市發展之經濟-生態效率。 關鍵詞：經濟-生態效率、資料包絡分析、Tobit迴歸分析 / Urban development creates economic values and results in environmental impact at the same time. Based on the concept of sustainable development, several methods, such as sustainable indicator frameworks, ecological footprints, urban metabolism, were reported to assess urban sustainability. However, these methods seldom focus on the relation between environment and economy. Therefore, World Business Council proposed Eco-efficiency. This research tried to apply Eco-efficiency to analyze which cities/counties are relatively eco-efficient and find causes bringing about discrepancies among twenty-two cities/counties in Taiwan. In the first stage, this research applies Data Envelopment Analysis(DEA) to figure out Eco-efficiency scores and applies Tobit regression method to analyze in the second stage. The results show that efficiency scores of northern cities/counties are higher than other areas and the ones of southern cities/counties are lower above four areas. Further discussion of analyzing from individual county/city perspectives, the efficiency scores of Taipei city, Hsinchu county, Hsinchu city, Taichung city, Chiayi city and Taitung county are relatively better in 1996, 2001 and 2006. Within comparatively inefficient cities/counties, oil consumption and electricity need to reduce. Water consumption and industry employees are superior factors in twenty-two cities/counties. Furthermore, the intensity of urban development and polluting industry ratio of second and third industries have significant positive effects on the eco-efficiency scores. Disposable income per person per year has significant positive effect on the eco-efficiency scores. Thus, this research suggests that government should improve the regulation on industry pollution. In addition, for those counties/cities with high intensity urban development need to ameliorate energy efficiency and reduce environmental impact. Key words：Eco-efficiency、Data Envelopment Analysis、Tobit regression

本研究發展一能夠預測學習者閱讀英語文章時之「個人化閱讀焦慮預測模型」，此預測模型係以資料探勘技術為基礎，透過資料探勘技術於個人閱讀歷程及標註行為中進行閱讀焦慮預測規則的建立，並將預測結果與判定規則回傳給教師，以提供教師掌握造成學習者閱讀焦慮之關鍵因素，並提供適當閱讀輔助策略，藉此減緩學習者閱讀焦慮程度，進而提升其閱讀學習成效。 為了驗證本研究所發展「個人化閱讀焦慮預測模型」的可用性，以及探討本研究所設計不同學習機制對閱讀焦慮減緩策略的有效性，本研究以準實驗法設計三種不同閱讀學習機制並分別實施於控制組、實驗組A與實驗組B，接著以臺北市立萬芳國中一年級學生作為實驗對象，進行本研究實驗資料的收集，以作為驗證「個人化閱讀焦慮預測模型」可用性的資料來源及三種學習機制間降低閱讀焦慮與提昇閱讀學習成效的有效性驗證資料。 研究結果發現，「個人化閱讀焦慮預測模型」能有效預測學習者閱讀焦慮程度，為一個可靠的閱讀焦慮程度判別工具。此外，本研究發現低焦慮學習者在閱讀標註互動上較高焦慮學習者使用頻率高，顯示高低焦慮程度學習者在閱讀標註互動行為上有較明顯的差異，而本研究在實驗組A所實施的合作式閱讀機制能有一定程度能降低中焦慮組學習者閱讀焦慮現象；在實驗組B所提供的線上教師閱讀輔助策略亦能有助於學習者閱讀焦慮減緩。除此之外，本研究所設計三種不同學習機制皆能有效提昇學習者閱讀學習成效，顯示本研究所發展合作式閱讀標註系統有助於提昇學習者閱讀學習成效。   最後，將研究結果進行整理同時輔以文獻驗證，並歸納研究者在研究過程中觀察發現，提出個人化閱讀焦慮預測模型修正、合作式閱讀標註學習社群與電子書閱讀輔助應用等未來研究議題的初步架構，供後續研究者參考以進行更深入的探討。 / To effectively reduce reading anxiety while reading English articles, this study employs C4.5 decision tree, which is a widely used data mining technique, to develop a Personalized Reading Anxiety Level Prediction Model (PRALPM) for learners based on individual learners’ reading annotation behavior on a digital reading annotation system. The proposed PRALPM can explore the key factors that cause reading anxiety based on the fired prediction rules determined by decision tree. Through understanding these key factors that cause reading anxiety, instructor can support appropriate reading strategies to reduce learner’s reading anxiety level and promote their reading performance. To assess whether the proposed PRALPM can effectively assist instructor to reduce reading anxiety, this study adopted the quasi-experimental method to compare the learning performances of three learning groups, which are respectively supported by a digital reading annotation system with different learning mechanisms for reducing reading learning. Among the three learning groups, the control group, experimental group A and experimental group B conducted the same English reading learning activity, but were respectively distributed a digital reading annotation system with individual annotation, cooperative annotation and cooperative annotation with instructor’s support based on the proposed PRALPM for reducing reading anxiety. The experiment were executed on Taipei Municipal Wan-fang High School 7-grade student, and collected experimental data for verified the model availability and the effectiveness of different learning mechanism in lower learner’s reading anxiety level. The results found that PRALPM can predict learner’s reading anxiety level efficacious, and it’s also a reliable tool for identify reading anxiety. In addition, the study found that low level anxiety learners has more reading interactive than high level learners, it also mean different anxiety level learners have Significant differences in reading interaction activities. And the Collaborative reading mechanism can help middle-anxiety-level learner reduce their anxiety efficacious in experimental group A. The online teacher reading assisted strategy can also help learners to slow their read anxiety in experimental group B. Furthermore, three type of learning mechanism all have the positive Effect to enhance learner’s reading performance, it shows that this collaborative reading annotation system can help learner Have better learning outcomes. At last, the study summarized the researchers observed and bring forward some future research issues such as PRALPM modify, cooperative learning community and the application of e-book reader-assisted subject.

由於電子化病歷逐漸取代了傳統的紙本病歷，在流通分享上面比傳統的紙本病歷更加來的方便及快速，另外電子病歷的整合性，也是比傳統的紙本來的有效。近年來雲端運算的發展，使得醫療系統在電子病歷上能夠更快速的發展，但是取而代之的是卻是雲端運算所產生隱私權的問題，在快速發展的雲端運算環境中，目前似乎無法完全確保資料的隱私性。即使現有的研究中可以讓資料擁有者表示自己的隱私偏好，卻因為設計時缺乏語意的考量，造成執行上有語意的落差。 本研究將探討電子病歷存放在雲端環境上，設計一套三層整合平台系統並使用語意化技術本體論整合來自多方的資料，達成在資料庫上使用OWL2作為整合的語言，並在此整合平台進行本體論整合，能夠讓使用者可以從多方的醫療中心快速查詢整合的資料，經由整合平台的改寫，到下層的規範擷取到上層平台進行管理與落實動作，最終在資料庫查詢資料，達成整合分享的目標，並同時能夠兼顧資料擁有者的隱私期待，完成在雲端環境上資料分享、整合、隱私保護的目標。 / The Electronic Health Records (EHRs) have replaced the traditional paper Health Records gradually and they are more rapid and more convenient in data sharing. Furthermore, the EHRs are also better than paper health records when health records need to be integrated on the computer. In recent years, the rapid development of cloud computing can help Health Information System to be more dynamic and provide a better service, but the problem of privacy is a critical issue. Although recent research can let data owner expresses his own personal privacy preference in to policy to protect privacy, it is lacked of semantics and that will result in the gap between the real meaning of personal privacy preference and of policy. In our research, we will using semantic technology to express personal privacy preference in to polices and also design the 3-layer integration platform to achieve semantics data integration so that polices can be enforced without loss of real meaning of personal privacy preference and polices will have interoperability with others when we are using semantic data integration.

近年來台灣連鎖咖啡店消費人口的穩定成長，提升了連鎖咖啡店的市場規模與消費產值，傳統利潤導向的市場經營方式，使得連鎖咖啡店的競爭更趨激烈，如何訂定正確的選址與經營策略，成為在高度競爭市場中存活的重要關鍵。 傳統的選址問題需要投入大量的人力與時間進行相關資訊的蒐集、訪查與評估，故而在新設營業點時，較少運用複雜的因素進行區位選址的分析與評估。因此能透過較多的因素，從區位選址與營利效應等觀點進行分析，協助投資者獲得更好的利潤，提高決策成功的機率，是極為重要的問題。 本論文的目的，在於為連鎖咖啡店之選址決策，提出能增加成功機率之設點建議。我們依據連鎖咖啡市場雙雄在訂定選址決策的成功經驗，透過相關係數進行人口與經濟活動因素之統計分析，以找出其成功選址之關鍵因素。同時運用資料探勘的分類技術，建構成功選址之分類模型，並經由地理資訊系統提供的圖層資料，對連鎖咖啡市場雙雄之競爭關係進行分析與評估，以提供正確選址及設點之建議。 實作中我們採用台北市出租店面之空間資料，以探討並評估本研究建議模型之實際效益。實驗結果顯示，透過本研究之選址分類模型進行設點類型之預測，有七成以上之達成率，顯示本研究提出之模型能有效增加選址的成功機率，同時經由競爭對手設點空間關係之分析，亦能提供有利選址決策之建議。 / The number of customers of coffee shop chains has grown steadily in recent years that cause the market size as well as the total consumption value increase rapidly and continuously. The competition among the chain coffee stores get even worse under the traditional profit oriented management style. In such case, it is crucial to make the correct decisions when selecting the coffee shop locations as well as making operation strategies in opening new coffee shops. Traditionally, it takes a great amount of time and human resources in collecting relevant information, conducting field visits as well as site evaluations when making coffee shop site selections. One seldom considers complex factors of site evaluation or field analyzing in selecting the location of new coffee shop. Hence, it will be one of the major contributions if one can find a mechanism in analyzing the site selection as well as profit evaluation to help the investors to produce better profit and to improve the chance of success. The goal of this thesis is to provide recommendations to improve the success rate of chain coffee shop site selection strategy. Based on the coffee market leaders’ success experiences in formulating the site selection strategies, we analyzed the correlation coefficients of the population as well as economy activities in order to identify the key factors in successful site selection strategies. We also used data mining techniques to construct the classification models of successful site selection. In addition, we analyzed and evaluated competition relations between the two leading chain coffee brands using the geographic information systems to obtain appropriate recommendations in new site selections. The shop rental information of Taipei City was used to explore and to evaluate the models recommended in our mechanism. The experimental results showed that the prediction through the classification models for site selections can achieve 70% of success rate. This indicates our mechanism effectively improve the successful rate of site selections. Moreover, the experimental results also show that the spatial analysis of site selections between the competitors is helpful in providing appropriate site selection strategies.

近年來保險經紀人產業蓬勃發展，保險業行銷通路多元化，其相關產業競爭環境更顯嚴苛險峻，而以「人」為本的人身保險產業更須著重良好顧客關係的維持，並滿足顧客之需求，以提升企業之形象與價值。因此能否有效應用資料採礦技術於顧客關係管理，精確地區別顧客族群並瞭解各自之需求所在，以整合各項通路資源進行目標市場行銷，並歸納出未續繳保費可能性較高之族群特性，作為顧客留置之用，減少新客源開發之成本，已然成為保險經紀人產業重視的課題。 本研究以T保險經紀人公司為例，說明如何應用資料採礦技術萃取出有利於個案公司顧客關係管理的知識，期間藉由個案公司提供的2005年至2010年保單資料作為實證分析之用。於分析過程中，使用敘述性統計以瞭解此六年期間新契約保單之概況；透過交叉分析知曉險種與被保險人、保單特性之關聯；利用被保險人和保單特性進行二階段集群分析，藉由分群結果作為市場區隔之用；最後使用被保險人、保單和業務員特性進行C&R Tree決策樹分析，以歸納出未續繳保費之分類規則。 根據本研究之發現，可將顧客分為理財退休規劃型、避險儲蓄需求型、人生風險規劃型和防癌醫療保障型四大顧客族群，可藉由目標市場族群特徵計畫行銷方案，以開發新客源創造利收；而利用C&R Tree決策樹分析可歸納出未續繳保費可能性較高的三項規則，藉此針對可能流失之顧客族群擬訂對應策略，以鞏固既有客戶提升續繳率；此外，於研究過程中亦發現個案公司資料庫缺在些許問題，應落實資料庫管理之正確性、即時性、完整性，以有效運用資料創造商業價值。期能達到目標市場行銷效益最大化、提升顧客留置策略有效性、改善當前資料庫管理之詬病。