毕业设计（论文）-网上教学平台的研究.doc
摘要摘 要随着Internet的飞速发展以及个人计算机的普及，网上教学已经成为最有价值、最有发展前景的行业，因此基于互联网的教学平台曾出不穷，现在广泛应用的中英文网上教学平台包括安琪尔(ANGEL)(CyberLearningLabsInc.,2003)、安龙(Anlon)(AnlonSystems Inc.,2001)、阿维拉网上导师(AvilarWebMentor)(Avilar TechnologiesInc.,2003)等等。然而纵观现有的网上教学平台，普遍存在着这样几个问题：网站结构纷繁复杂，层次不清晰，目标性不强；教学形式单一乏味，不能很好的调动学生的学习兴趣；教学针对性差，不同学习能力、不同基础的学员均采用同样的教学方法等等。这些问题的存在在不同程度上影响了网上教学平台的的教学成果，同时也制约了远程网络教学的发展。本论文的课题研究并建立一套大型嵌入式系统网上智能教学平台，能够实现网上教学、网上答疑、网上交流、网上测试等。该教学平台包括大量嵌入式系统领域的网上教学课件、资料、案例等，通过网上教学方式实现对复合型、应用型嵌入式人才的培养，对于国内外的网络教学具有十分重要的意义。研究聚类分析算法的概念、原理、分类等相关技术，着重研究了K-均值聚类分析算法和网格聚类算法，以及以国内外研究的现状，提出一种新的GBKM聚类分析算法，该算法整合了网格聚类与K-均值聚类，并且提出了一种新的划分网格的算法、以及新的计算密度阀值的函数，经过理论分析以及试验证明，论证了GBKM算法的优越性，聚类过程达到了令人满意的效果。4：阐述了“个性化智能”教学的概念，并参与实现了嵌入式网上智能教学平台；该平台涵盖了网上教学、网上答疑、网上交流、网上测试等十四大服务，功能强大，服务周到。论文期间，本人所做的模块主要有网站论坛、精品课程、资源中心等等。5：详细说明了GBKM聚类分析算法在网上智能教学平台中的应用，设计了个性化智能学习系统，将GBKM聚类分析算法作为成绩分析的核心算法，通过将学员成绩记录进行聚类分析得到学员的个性特征，个性化智能学习系统根据分析结果对学员进行个性化的学习指导，取得了令人满意的“个性化智能”教学效果。 6：最后总结本次研究所取得的成果，并提出不足与待改进的部分，供以后继续研究。关键词：嵌入式； 智能教学； 聚类分析；K-均值算法75目录AbstractWith the rapid development of Internet and the popularity of personal computers, the Internet has become the most valuable teaching, the most promising industries, Internet-based teaching platform has been a non-poor, now part of the wider use of English in teaching online platform including firsthand (ANGEL) (CyberLearningLabsInc., 2003), Anlong (Anlon) (AnlonSystems Inc., 2001), Avila online instructors (AvilarWebMentor) (Avilar TechnologiesInc., 2003) and so on. However Looking at the existing online teaching platform, widespread following questions: Site complicated, levels not clear targets are not appropriate; Teaching form a single boring, and not very good mobilize students to learn; Teaching targeted poor, different learning ability, based on different school Members are using the same teaching methods, and so on.The project to establish a large-scale intelligent embedded systems online teaching platform can be achieved online teaching, online FAQ, online exchanges, and online testing.The teaching platform included a large number of embedded system fields online courseware, data, case, etc., through online teaching methods to achieve the complex, application-embedded train professionals. While in-depth study of data mining association rules and clustering analysis, and other advanced technologies, computational intelligence,network security,Internet expert system in the application of this system improved the intelligence. This article reads as follows: A: To study the status of remote network teaching, technology and the existing problems, proposed a "personalized smart teaching" concept, the research projects in the background necessary to do the summing up and made a study on this issue of great significance. Second: The cluster analysis algorithm concept, principle, classification, and other related technologies, focusing on the K-means clustering algorithm and analysis grid clustering algorithm, the algorithm for future improvement for full preparations. 3: detailed analysis of the K-means clustering algorithm and grid clustering algorithm research and the improvement of the past, and based on this, a new GBKM cluster analysis algorithm, the theoretical analysis and tests proved that the improved algorithm to achieve satisfactory results. 4: Details on Embedded Intelligent online teaching platform used by the building and the key techniques to enhance their websites smart, reuse, and robustness. 5: GBKM described in detail in the cluster analysis algorithm intelligent online platform of teaching - personalized intelligent learning system integration applications, and achieved satisfactory "personalized smart" teaching effect.6: The final summing up the results of the Institute, and proposed to be improved with insufficient part for the future to continue to study. Key words: Embedded; Intelligent Tutoring; Cluster analysis; K-means algorithm目 录摘 要IAbstractII目 录III绪 论1第一章 网上教学平台的研究41.1 远程网络教学的概述41.1.1 远程网络教学的概念与发展41.1.2 远程网络教学的显著特点41.1.3 远程网络教学的基本模式51.1.4 远程网络教学存在的问题和“个性化智能教学”的提出71.2 网上教学平台的技术与理论研究81.2.1 应用于网上教学平台的流行技术81.2.2 网上教学平台的结构101.2.3 网上教学平台建设的基本原则11本章小结12第二章 聚类分析算法的研究142.1 引言142.2 聚类分析的简介152.2.1 聚类分析的概念152.2.2 聚类分析的原理152.2.3 聚类分析中的数据类型162.2.4 相似性测度172.3 聚类分析的算法和技术182.3.1 聚类方法的分类182.3.2 几种典型的聚类方法192.3.3 算法的比较212.4 K-均值聚类与网格聚类212.4.1 K-均值聚类（K-means Clustering Method）212.4.2 网格聚类（Grid-Based Clustering Method）23本章小结24第三章 聚类分析算法的改进253.1 引言253.2 改进的算法253.2.1 算法的基本概念253.2.2 算法的基本思想263.2.3 算法的步骤273.2.4 算法的流程图273.2.5 算法的实现293.3 改进算法的性能验证323.3.1 时间复杂度分析323.3.2 试验结果对比323.3.3 纯度比较363.3.4 凝聚度和分离度评价37本章小结39第四章 嵌入式网上智能教学平台的设计404.1引言404.2 开发环境、工具以及相关技术414.2.1 开发环境与工具414.2.2 相关技术414.3 系统的开发模型434.3.1 模型设计434.3.2 模型实现454.4系统分析与功能设计474.4.1 系统分析474.4.2 模块设计484.4.3 系统预览504.5 数据库设计534.5.1 数据库表的设计534.5.2
