一般數位學習系統中，均缺乏迅速有效學習類型分類機制，在此種環境下，若欲考量不同學習類型以提供學習者適性化教材，首先須大量地蒐集學習者的學習特徵資料，再加以分析分類，此一過程往往需耗費許多時間。本研究將資料探勘技術中的k個最鄰近點（k - nearest neighbor）分類法、配合基因演算法應用於學習者管理平台上，將不同學習類型分類，並篩選出可快速辨識學習類型的學習者行為特徵，本研究之目的在於建立可自我調適(self-adjustment)、快速有效學習類型分類機制，可做為學習管理平台提供不同類型學習者適性化教材之依據。

The existing learning management systems (LMS) are in general not equipped with learning style classification facility. In such a situation, if a system intends to provide with learner-adaptive learning contents, it must collects enormous learning behavior data of learners and then process them with complicated computing. This task is usually time-consuming and needs lots of computing resources. In this paper we describe a fast, efficient and adaptive learning-style classification mechanism that is based on k - nearest neighbor classification (K-NN) approach and genetic algorithms. The contents presented in this paper can be used for reference for constructing LMS with adaptive learning contents providing.

參考文獻
[1] 丁一賢,陳牧言, “資料探勘,” 滄海書局.
[2] 林奇賢,“網路學習環境的設計與應用,”資訊與教育, 第67期, 1998.
[3] 林清賀,“網路學習系統上之學習歷程評量,” 國立中央大學資訊工程研究所碩士論文, 1998.
[4] 吳弘凱,“國小學童數位學習擷取課程行為樣式分析,” 國立台南大學資訊教育研究所碩士論文, 2004.
[5] 邱瓊慧,朱治平,“學習者學習行為分析研究,” 資訊工業策進會分包學術機構研究計畫期末報告, 2003.
[6] 陳奕錡,“符合SCORM規格具有策略推薦之策略編輯工具設計與實作,” 國立成功大學資訊工程研究所碩士論文, 2005.
[7] 張瑞程,“支援學習策略選取之智慧型規則管理機制之研究,” 國立成功大學資訊工程研究所碩士論文, 2005.
[8] 黃履峰,邱貴發,“學習歷程檢視器：Web學習環境中的學習管理工具,” 第八屆電腦輔助教學國際研討會大會論文, 台中市, 逢甲大學, 1999.
[9] 鄭安授,“電子報使用者遊歷行為之描繪-以交大學生為例,” 國立交通大學傳播所碩士論文, 2001.
[10] 蔡旻芳,“網路學習之學習歷程分析系統,” 國立中山大學資訊管理研究所碩士論文, 2000.
[11] 劉鼎康,“應用類神經網路進行垃圾郵件過濾之研究,” 中原大學資訊管理系碩士論文,2005.
[12] Advanced Distributed Learning, SCORM Overview, 2004.
[13] Advanced Distributed Learning, SCORM Content Aggregation Model, 2004.
[14] Advanced Distributed Learning, SCORM Run-Time Environment, 2004.
[15] Advanced Distributed Learning, SCORM Sequencing and Navigation, 2004.
[16] A.Whitney, “A direct method of nonparametric measurement selection,” IEEE Trans. Computer., vol.20, pp.1100-1103, 1971.
[17] D.L. Wilson, “Asymptotic properties of nearest neighbor rules using edited data,”IEEE Transactions on Systems, Man, and Cybernetics, pp. 408-421, 1972.
[18] Jian-Hung Che、Zhao-Hong Yen、Shinn-Ying Ho, “Design of Optimal Nearest Neighbor Classifier Using an Intelligent Multi- Objective Evolutionary Algorithm,” Lecture Notes in Computer Science (LNCS), vol.3157, pp.262-271, Springer, 2004.
[19] Kuncheva, L. I. and Jain, L. C., “Nearest neighbor classifier: Simultaneous editing and feature selection,” Pattern Recognition Letters, vol.20, pp.1149-1156, 1999.
[20] L. I. Kuncheva, “Editing for the k-nearest neighbors rule by a genetic algorithm,” Pattern Recognition Letter, Vol.16, pp. 809-814, 1995.
[21] L. I. Kuncheva, “Fitness functions in editing k-NN reference set by genetic algorithms,” Pattern Recognition, Vol.30, No.6, pp. 1041-1049, 1997.
[22] P.E Hart, “The condensed nearest neighbor rule,” IEEE Transactions on Information Theory, Vo1.16, pp.515-516, 1968.
[23] S. Stearns, “On selecting features for pattern classifiers,” In 3rd International Conference on Pattern Recognition, Coronado, CA, pp. 71-75, 1976.
[24] S.-Y. Ho, L.-S. Shu and H.-M. Chen, Intelligent genetic algorithm with a new intelligent crossover using orthogonal arrays,” Proc. of the Genetic and Evolutionary Computation Conference, pp.289-296, 1999.
[25] Tom M. Mitchell, “Machine Learning,” McGRAW-Hill International Editions.
[26] Yu,P.T.,Own,C.M.&Lin,L.W., “On the learning behavior analysis of web based interactive environment,” International Conference on Computer in Education（ICCE2002）, USA, 2002, January.