近年來網路學習隨著網路的發展，逐漸受到重視。網路是一個開放式環境，擁有豐富的學習資源，並且突破學習時的空間限制、時間限制，讓使用者擁有較高的學習自主性。由於在網路學習環境中，學習者必須負擔選擇和學習的責任，容易有學習迷失的問題。因此，蘇俊銘等提出一個動態產生適性化教材順序的Instructional Activity Model(IAM)。更依據所提出的IAM，設計一個符合美國國防部所提出之SCORM標準的系統。 根據奧蘇伯爾在認知同化學習論中所提到，能力之間是會互相影響的。然而IAM並未考量能力與能力之間的相互影響關係。因此本研究將將結合奧蘇伯爾同化論，提出一個考慮能力與能力之間相互關係的適性化教學模型，提供學習路徑導引。期望藉由此模型，能夠提供E-learning環境設計者，設計更符合學習者能力狀況的適性化學習課程。 / E-Learning issues have been discussed and investigated recent years. Internet environment provides multiple choices of learning time and learning materials. Much research has been done on E-Learning. The standards of E-Learning platform have been developed to make the learning content sharable and reusable. Some research has been devoted to the design of adaptive Learning system for need of different users. The Instructional Activity Model (IAM) is a general-purpose model to generate an adaptive learning course, which is compatible with the SCORM standard. IAM is composed of related Activity Tree (AT tree) nodes and capability nodes. Prerequisites are capabilities supposed to posses before learning an AT tree while contributions are capabilities after learning an AT tree. IAM model supports the adaptive learning sequencing by considering the relationships between AT trees and capability nodes. However, the IAM model does not take the influence between capabilities into consideration, especially for the assimilation of capabilities. In this paper, we propose the mechanism to integrate the concept of assimilation theory to the IAM model. In our proposed mechanism, the relationships between capabilities are considered based on the similarity measure between capabilities. The selection process of IAM is also modified to reflect the relationships of capabilities. Simulation result shows that the proposed mechanism is helpful for the learning sequencing.

資訊科技對學習引發了全面性的變革，網路無遠弗屆、快速便利的特性，更使得網路學習、線上學習、數位學習等利用網路與資訊科技的結合的學習方式為大家所重視。未來邁入知識經濟後，人們更需要不斷的學習來提昇自己的競爭力，透過學習獲取知識。資訊科技與網路的結合，為數位學習帶來許多的優勢與好處，但是雖然數位學習的網站很多，大多數仍無法達到如傳統學習的學習效果，學習者通常瀏覽過一兩次後，便失去學習的興趣，因此數位學習系統仍缺乏整合性的規劃與開發。 　　目前網路上許多的數位學習系統提供的互動機制不足，學習者缺乏人際互動的工具與活動，除此之外，許多互動工具無法得到參與學習的使用者之青睞，學習者甚少使用，如同虛設。因此本篇論文企圖設計出一套整合性與互動的數位學習之設計，解決目前網路學習在互動中所遭遇的瓶頸。 　　本研究從適性學習(Adaptive Learning)、合作學習(Collaborative and Cooperative Learning)與建構學習(Constructive Learning)三方面進行設計，互動機制可以達到此三個目的，進一步讓系統可以支援達成合作學習與建構學習的活動設計。 / Information technology and Internet technology have brought the revolution of learning, training and education. The knowledge economy in the 21st century has made the knowledge become the most important critical success factor. We must continuously learn in order to maintain the competitive advantage. Many corporations and educational institutions have developed their own e-Learning web sites. Because the lack of adequate interactive mechanisms, most of them cannot provide the efficient and effective learning results. Our research aims to develop an interaction-centric e-Learning design and implementation. Based on the adaptive learning, collaborative and cooperative learning, constructive learning concepts, we design an integrated interaction-centric model to develop a set of toolbox to enhance the interactive activities of e-Learning.

本研究的目的在建構具備智慧型媒體特性之網路教學系統以增進網路學習的效果，特別是如何輔助學習者理解空間關係，以增進對空間幾何座標轉換的學習成效。電腦運算能力的增進使其成為極具潛力的教育媒體。基於教學及教育媒體的理論，本研究倡議在典型的網路教學中加入「智慧型媒體」的考量與設計以增進學習成效。智慧型媒體包括了兩個面向的考量，我們稱之為「媒體與方法」的考量 (Media and Method concern)。從媒體的角度來考量，電腦多媒體的使用應以能增進學習效果為原則，從「認知媒體」(Cognitive Media)的角度出發來設計網路教學的媒體呈現方式，媒體的目的在於清楚地傳達領域之知識給學習者。而從方法的角度來考量，應考慮如何運用電腦運算的特性以實現其他教育媒體裝置不能實現的教學策略及方法，例如互動式及適性化的教學。接續過去「智慧型教學系統」(Intelligent Tutoring System)及「適性化超媒體」(Adaptive Hypermedia) 的研究，本研究提出一套適性化的機制，將一般適性化系統中課程排序(adaptive course sequencing)的機制明確分離為「學習概念排序」及「教材選擇」兩個部分，達到更佳的抽象化及運用教學策略上的彈性。 本研究以「空間座標轉換」做為領域知識，基於「媒體與方法」的考量，設計了稱為CooTutor (Coordinate Tutor)的網路教學系統來輔助空間座標轉換的學習。運用了電腦動畫技術，這個系統使用互動式三維媒體(Interactive 3D Media)清楚有效地傳達領域知識。由於空間座標轉換的學習相信與學習者的空間能力(Spatial Ability)相關，本研究透過實驗來探討互動式三維媒體的使用與空間能力增進之間的關係。另外，我們也研究並評估如何將空間能力及學習風格(learning styles)等學習者個人特質作為適性化依據，以及如何設計相對應的適性化機制。 本研究的主要貢獻包括了 (1) 提出了使用智慧型媒體的概念，以「媒體與方法」的考量來討論網路教學的學習成效，及 (2) 提出一個創新及可行的架構將互動式三維媒體及適性化技術結合、運用於網路學習的學習模式上以輔助空間座標轉換的學習。 / The objective of this research is on developing a Web-based educational system with intelligent media to enhance learners’ learning effects, especially to facili-tate learners’ spatial reasoning on learning spatial geometry topics. The increas-ing computing power allows us to use computers as powerful educational media. Based on theories of pedagogy and educational media, we propose to integrate intelligent media into typical Web-based learning paradigm to improve learning. “Intelligent media” in this research refers to two aspects of considerations. They are media—cognitive media aspect and method—intelligent tutoring aspects. The consideration of cognitive media aims at offering learners the most ease-of-understand presentation of a particular domain. The consideration of in-telligent tutoring targets to offer learners personalized learning experience based on individuals’ learning needs. To achieve better abstraction and flexibility in the adaptive mechanism, we have chosen to separate the concept sequencing from the underlying task of selecting appropriate learning materials. By considering the characteristics of spatial geometry concepts, a Web-based learning environment called CooTutor (Coordinate Tutor) for learning spatial geometric transformation (SGT) is developed. Interactive 3D media is integrated into the system for delivering domain concepts effectively. Since the domain, spatial geometric transformation is evidently related to spatial ability (a group of human abilities about the use of space). This research attempts to address the relation between spatial ability and interactive 3D media via experimental evaluations. Moreover, learners’ latent traits, including spatial ability and learn-ing styles are considered to be used in adaptive material selection. The main contribution of this research would be (1) the conceptualization of in-telligent media and the M&M concern for effective Web-based learning, and (2) an innovative approach and tenable architecture of employing 3D computer graphics and adaptive technologies in Web-based learning context for SGT learning.

本論文主要在估計門檻式自動迴歸模型之參數的信賴區間。由線性自動迴歸 模型衍生出來的非線性自動迴歸模型中，門檻式自動迴歸模型是其中一種經常會被應用到的模型。雖然，門檻式自動迴歸模型之參數的漸近理論已經發展了許多；但是，相較於大樣本理論，有限樣本下參數的性質討論則較少。對於有限樣本的研究，Woodroofe (1989) 提出一種近似法：非常弱近似法。 Woodroofe 和 Coad (1997) 則利用此方法去架構一適性化線性模型之參數的修正信賴區間。Weng 和 Woodroofe (2006) 則將此近似法應用於線性自動迴歸模型。這個方法的應用始於定義一近似樞紐量，接著利用此方法找出近似樞紐量的近似期望值及近似變異數，並對此近似樞紐量標準化，則標準化後的樞紐量將近似於標準常態分配，因此得以架構參數的修正信賴區間。而在線性自動迴歸模型下，利用非常弱展開所導出的近似期望值及近似變異數僅會與一階動差及二階動差的微分有關。因此，本論文的研究目的就是在樣本數為適當的情況下，將線性自動迴歸模型的結果運用於門檻式自動迴歸模型。由於大部分門檻式自動迴歸模型的動差並無明確之形式；因此，本研究採用蒙地卡羅法及插分法去近似其動差及微分。最後，以第一階門檻式自動迴歸模型去配適美國的國內生產總值資料。 / Threshold autoregressive (TAR) models are popular nonlinear extension of the linear autoregressive (AR) models. Though many have developed the asymptotic theory for parameter estimates in the TAR models, there have been less studies about the finite sample properties. Woodroofe (1989) and Woodroofe and Coad (1997) developed a very weak approximation and used it to construct corrected confidence sets for parameters in an adaptive linear model. This approximation was further developed by Woodroofe and Coad (1999) and Weng and Woodroofe (2006), who derived the corrected confidence sets for parameters in the AR(p) models and other adaptive models. This approach starts with an approximate pivot, and employs the very weak expansions to determine the mean and variance corrections of the pivot. Then, the renormalized pivot is used to form corrected confidence sets. The correction terms have simple forms, and for AR(p) models it involves only the first two moments of the process and the derivatives of these moments. However, for TAR models the analytic forms for moments are known only in some cases when the autoregression function has special structures. The goal of this research is to extend the very weak method to the TAR models to form corrected confidence sets when sample size is moderate. We propose using the difference quotient method and Monte Carlo simulations to approximate the derivatives. Some simulation studies are provided to assess the accuracy of the method. Then, we apply the approach to a real U.S. GDP data.