在邁入知識經濟的現今社會中，企業的生產要素已轉變為以知識為基礎，而學習正是知識經濟時代增加企業競爭力的最佳利器。隨著科技的發展，線上學習(e-Learning)不論在傳統公司或電子化企業中已扮演著重要不可或缺的角色。線上學習的獨特性不但可使教育訓練的成本降低，更可以配合企業的全球化運作，這些優點吸引了許多公司開始規劃線上訓練。但由於目前線上學習多採用Web(World Wide Web)介面並且必須與網際網路連接，所以有發展上的限制且缺乏行動性。因此本研究將應用行動科技結合學習科技來克服這個問題。 　　本研究希望建構一行動學習設計並且整合知識庫，使得知識學習的來源不再僅侷限於學習教材，更能包含整個組織的知識。另外也設計一搜尋機制希望藉此能有效率地執行行動搜尋。 / In knowledge economy, the key factor of production is knowledge. Learning will be the best effective tool to increase the competitiveness in the knowledge-based economy. With the development of Information Technology, Electronic Learning (e-Learning) becomes important, indispensable in traditional business, and e-business. e-Learning with unique characteristics lowers training costs and makes organizational learning global. These benefits catch the attentions of the company to pursue e-Learning. However, the problem that e-Learning has to be Web-enabled (World Wide Web enabled) and Internet-based creates a technical and managerial barrier limits its development and lack the mobility. In our research, we aim to apply Mobile Technology with Learning Technology to overcome this problem. 　　Our research focuses on the design of a mobile learning (m-Learning) model that supports learning anywhere, anytime and any form. Beside, we develop an approach to integrate with knowledge repository. The learning scope will not be limited to instructional courseware. It includes organizational knowledge. The search assistant helps user to access relevant information with minimal effort and reduce the amount of information that needs to be transmitted.

由於台灣網路ICP新聞資料庫已被消費者大量搜尋、閱讀與使用，本研究選擇提供網路新聞內容查詢為主要服務項目的聯合知識庫為對象,進行實證探討。在聯合知識庫的網站的首頁上，針對網路族群為樣本對象，將問卷置放於網路環境以供填答。總共回收2817份有效樣本。 本研究應用G.B.Sproles 與E.L.Kendall(1986)兩位學者之論文發展出消費者決策特性的量表；同時為了測量消費者對服務品質的重視程度，依照了 Parasuraman、Zeithaml、及Berry(1988)三位學者所建構的SERVQUAL量表來做為ICP內容查詢服務特性的20個變項的尺度；然後再選擇廣為採用的四種線上收費模式來進行偏好的實證分析。而整個研究架構的支撐是以消費行為學者Del.I. Hawkins, Roger J. Best & Kenneth A. Coney(1992)所發展的行為研究模式來建立。期望能於研究結果發現網路族群心中最偏好的收費方式。 研究所使用的統計分析方法則有因素分析、T檢定、卡方考驗、變異數分析、迴歸分析等技術，藉以探討網路族群對查詢新聞資料的服務品質重視、消費決策特性、與收費模式偏好之間的相關差異與效果。整體研究結果可作為未來行銷策略的擬定參考。 研究結果發現： 1. 網路族群對ICP內容新聞資料庫收費模式偏好有顯著差異。 2. 網路族群對ICP內容新聞資料庫服務品質所重要的程度不同,對收費模式偏好有顯著差異。 3. 不同網路消費決策特性對收費模式偏好有不同看法。 4. 不同網路消費決策特性所偏好的收費模式,在服務品質重視度上的看法有顯著差異。 5. 年齡與教育程度是影響收費模式偏好之重要因素。 6. 人口統計變數之不同，會影響樣本對小額付費制與預付制之偏好程度。網路族群的人口統計變數對收費模式偏好的顯著相關之中，對小額付費制與預付制有明顯的看法。不同的個人屬性會有不同偏好。其中只有年齡與教育程度有明顯相關，年齡越輕越偏好小額付費,年齡越長(多於年輕者)則偏好預付制；而教育程度高的人有部份會偏好小額制。其它的變項則無明顯差異。 7. 網路族群可接受新聞內容查詢資料庫多元化設計的收費模式。 8. 網路族群較偏好預付制收費模式並期望所享折扣應較多。 9. 因素分析後的三大服務品質因子與四大消費決策特性對收費模式皆呈顯著相關,收費模式對上述二類因子也呈現部份的差異性。 10. 網路族群對ICP內容新聞資料庫服務品質重要性的重視程度經因素分析後，分別為：重視內容品質、使用互動便利性、即時時效性等三大因子。 11. 網路族群在ICP新聞內容查詢服務的消費決策類型可歸納為：理性經濟型、感性衝動型、低價傾向型及基本功能接受型四種類型。 12. 樣本最偏好固定會費者,對服務品質沒有強烈偏好,比較重視內容。最偏好預付制的人最重視內容品質。最偏好小額付費的人最不重視新聞即時性。他們也是最不偏好預付制的人。內容品質及即時性都對小額付費的人不重要。 13. 本研究得出兩個收費模式偏好預測模式，可預測消費者選取固定年費制與預付制之機率。兩個收費模式偏好預測模型，分別如下： logit(p1)=log[p1/(1-p1)]=0.1448*X1+0.1381*X2 p1為選取固定年費之機率 X1為便利性平均重視程度之標準化分數 X2為相關性平均重視程度之標準化分數 logit(p2)=log[p2/(1-p2)]=0.1198*X1+0.1*X2 p2為選取預付制收費之機率 X1為可靠性平均重視程度之標準化分數 X2為充分性平均重視程度之標準化分數 本研究同時發現：現行於ICP新聞內容資料庫查詢服務的收費方式,在實務界是採單一的收費模式，此種方式並非消費者最偏好的選擇。網路族群會因服務品質的重視度不同而有不同偏好。針對於不同族群，網路經營者對收費模式的擬定，未來應以不同的個人屬性而訂定，依循其所重視的服務品質面向，提供多元而彈性的收費策略，以提昇網站的競爭力。 關鍵字：ICP新聞內容查詢、服務品質、消費決策特性、收費模式偏好 / The Analysis on the Internet Consumers' Decision-Making Behavior and their Preferences to the Service Quality and Pricing Models in News-related Online Database & A Case Study on UDNDATA.com Adviser: Dr. Chu Fongkang Tung-Ju Yang ABSCRACT As consumers vastly search, read and use Taiwan's Internet news and database,the research purpose is to exam the online news content search service, udndata. com. The first dimension of the research is based on the models of consumer’s decision-making behaviors which were developed by G.B. Sproles and E.L. Kendall (1986). The other dimension is based on the models from the SERVQUAL scale by Parasuraman、Zeithaml and Berry. The scale was constructed in 1988 and was proved effective. The research use the consumer’s behavior model invented by Del. I Hawkins , Roger J. Best & Kenneth A. Coney in 1992. The research's questionnaireswas posted on udndata.com which randomly allowed average internet consumers to complete. The first dimension of the questionnaire is using ICP content search service's 20 variables to measure the degrees of consumers' regards; the second dimension is about four available online pricing models and let consumers select their own preference model; the third dimension is to look into online consumers' buying behaviors. The methodology of the research is including descriptive statistics, factory analysis, chi-square, ANOVA and regression analysis, to exam online consumers' regards, consuming behaviors and pricing preferences toward search news data service. The findings of the research can be references to future marketing strategies. There are 2,817 effective samples over the udndata.com, and the findings are as below: 1. For online users, there are significant differences to the content search pricing models. 2. For online users, the different degrees of regards to content search quality service will significantly affect their preferences toward pricing models. 3. Different online consuming decision-making behaviors will affect their preferences toward pricing models. 4. For on-line users , different demographics will affect the preferences toward pricing models. The variables of age and education will affect the preferences toward the micro payment and prepaid payment. The youths and high education status prefer the micro payment and the old prefer the prepaid payment. 5. There are significant differences to the pricing models' preference of different online consuming characteristics and the degrees of regards to content service quality. 6. For online users, it is possible for them to choose the multiple pricing models which are close to their personal demographic characteristics. 7. For online users, they prefer to the prepaid pricing model and expect more discounts choices. 8. The 3 groups of the quality services and 4 groups of the consumers’ decision-making behaviors through the factor analysis significantly affect the preferences toward the pricing models. 9. Through factors’ analysis, the online users put the emphasis on 3 characteristics as follows: 1.content quality 2.the convenience of interaction with users 3.the real time effect. 10. After factors’ analysis, the online users of the content search can be categorized as follows: 1. the economic type 2.the impulsive type 3.the low-price oriented type 4.the basic need type. 11. The people prefer the fixed-payment subscription are without strong liking to quality service, however, they pay more attention to the content. The people who prefer the prepaid payment concern the content quality more than others. The people who pay the least on the instant services prefer the micro-payment. The research also finds that the available pricing practice of ICP news content search is single and fixed, it's not consumers’ first choice. Because the online consumers' preferences are varied by their different degrees of regards. In future, the online marketers should emphasize more on different humanity attributes to customize pricing models for different consumer groups, and also take good care of their first service dimension priority to offer more multiple and flexible pricing strategies to enhance their online competitiveness. Key Words: Internet Content Providers’ Service, Pricing Model, News Database, Consumer’ Decision-Making Behavior

知識管理在二十一世紀已成為世界的潮流，深深影響了人類的經濟活動，組織知識亦躍升成為企業內最重要的資產，在全球化的趨勢中為企業提供競爭力的泉源。現今大部份的知識管理系統研究，均從企業管理的角度，針對知識管理模型、知識管理成功因素等問題進行探討，但對於製造工廠基層中實際存在的核心知識管理方法，卻未見有相關的研究加以探討。在製造業中，企業最寶貴的工程製造技術知識是散佈在生產製造流程的各個工作程序與人員中，企業必須有系統的收集彙總後，始能有效運用、發展與持續企業的競爭力。 / 知識管理重在知識的分享，本研究建立的「製程整合知識庫」為知識的分享建立良好的基礎，對於新晶圓廠建廠生命週期運轉階段的製程技術移轉工作亦能提供協助，利用相似性移轉的方法，將「製程整合知識庫」完整移轉至新晶圓廠，可以提高製程的穩定度、提昇產品的良率，達到穩定製程的移轉安裝工作、縮短新晶圓廠量產運轉所需時間的目的。此外，對於營運中的工廠而言，「製程整合知識庫」也可協助檢視生產線上的缺失，以提昇產品品質與生產效率。