隨著電腦技術及網際網?的蓬勃發展，?位學習已成為當今教與學的另一種趨勢。以網?為基礎的課程順序應該因人而?，能動態安排課程順序以滿足個人化之需求是一重要的技術。因此，近??有許多研究者著手於發展適性化學習系統，期許能提供個人化的學習?徑；然而大部份的系統僅考慮學習者的興趣、喜好及瀏覽?為，而忽?學習課程的難??及先後備知?關係，?適當的課程常造成學習者的負擔及學習上之迷失，且?低學習成效。隨著人工智慧技術的快速發展，本體知??(ontology)能夠應用?表達課程概?間的學習概?關?，因此可以用?協助個人化學習?徑的建構。因此本研究在同時考?教材的難??、先後備知?關係及學習者能?的前提下，提出一個以本體知?概?圖(ontology-based concept map)為基礎之個人化學習?徑產生方法。透過自動產生建構ontology的技術，我們能夠以概?圖的方式呈現某一特定?域的重要知?及核心概?關係，據此所產生的學習?徑，可以有效?低學習迷思的問題及認知負載的發生，且可促進學習成效。本研究以國小分?單元作為實驗教材內容，透過資?探勘的方式，將國小分?單元彼此間的關?探勘出?，並以概?圖方式表示，加上配合基因演算法運用在個人化的學習?徑的規劃上，以期引導個人進?符合個人需求之有效?學習。 關鍵字：本體知?為基礎之概?圖、個人化學習?徑、智慧型教學系統、網?學習 Abstract Developing personalized Web-based learning systems has been an important research issue in the e-learning field because no fixed learning pathway will be appropriate for all learners. However, the current most web-based learning platforms with personalized curriculum sequencing tend to emphasize the learners’ preferences and interests for the personalized learning services, but they fail to consider difficulty levels of course materials, learning order of prior and posterior knowledge, and learners’ abilities while constructing a personalized learning path. As a result, these ignored factors easily lead to generating poor quality learning paths. Generally, learners could generate cognitive overload or fall into cognitive disorientation due to inappropriate curriculum sequencing during learning processes, thus reducing learning effect. With advancement of the artificial intelligence technologies, ontology technologies enable a linguistic infrastructure to represent concept relationships between courseware. Ontology can be served as a structured knowledge representation scheme, which can assist the construction of personalized learning path. Therefore, this study proposes a novel genetic-based curriculum sequencing scheme based on a generated ontology-based concept map, which can be automatically constructed by a large amount of learners’ pre-test results, to plan appropriate learning paths for individual learners. The experimental results indicated that the proposed approach is indeed capable of creating learning paths with high quality for individual learners to greatly reduce their cognitive overloads and to help learners learn more effectively.
