【作者】 张婧；

【导师】 杜江洪；

【作者基本信息】 华东交通大学 ， 计算机应用技术， 2009， 硕士

【摘要】 随着Internet的发展,网络教学得到了越来越广泛的应用,人们可以不受时间和空间的限制自主地参加网上学习。这种学习方式能够激发学习者的学习兴趣,从而达到让学习者主动构建知识的目的,实现自己获取知识、甚至创新知识的理想目标。因此,网络教学系统也成为了国内外教育技术领域的研究热点。但是目前的网络教学系统还存在着缺乏先进的学习理论的指导、系统设计缺乏智能性等缺点。论文针对现有网络教学系统存在的不足,引用Agent技术,设计了一个基于多Agent的网络教学系统模型。该模型分三个模块,各个模块由多个Agent组成。在论文中分别对这些模块的结构和功能进行了分析。模型的设计思想即发挥了教师的主导作用,又充分体现了学生的主体作用。系统的智能化设计方面,在知识表示方法上,为了对学科内容提供全局的认识,采用树结构来组织课程内容知识点,通过对知识点属性赋值建立知识点之间的语义联系,为有效控制教学序列提供依据。提出了基于错误诊断的强化学习策略,通过分析测试答案找出学生的知识缺陷,结合知识表示模型,推导出学生需要加强的合理学习路径,达到能对学生进行有针对性的强化教学的目的。提出了能够根据学生的认知水平自适应的进行组卷的智能组卷算法。针对现有学生模型的不足,构造了一个基于认知理论的学生模型,提出了评估认知水平、记录学习进度、估算学习偏好的方法。通过对多Agent和J2EE技术的深入学习和研究,提出了集成JADE和J2EE来实现智能网络教学模型的思想。该系统模型在一定程度上实现了智能化、个性化教学,可以为学习者提供更方便和更高效的网络学习环境。最后以《Java程序设计》课程为例给出了该系统的实验结果,验证了该系统的有效性和实用性。更多还原

【Abstract】 With the development of Internet, network teaching has been getting more and more extensive application, students can participate in online learning free and not restricting by time and space. This learning means can inspire learners to learn, so that learners can construct knowledge depending on their own initiatively and get, even innovate knowledge by themselves. As a result, the network teaching system has become the the research heat of educational technology field at home and abroad. However, there are still many shortcomings in the network teaching system nowadays, such as lack guidance of the advanced learning theory and intelligence.Aimed at defects of the present network teaching system, I design an intelligent network teaching system, which is based on Agent technology. The model is divided into three modules, each module consists of a number of Agent. In this thesis, I analyze the structure and function of each module. The designing concept of the model not only bring a leading role of teacher into play but also fully reflects the main body of student.In the aspects of designing intelligent system, in aspect of knowledge expression, in order to offer the overall view of discipline content, the tree structure is used to organize knowledge points, by setting some knowledge node attribute the semantic relation between knowledge nodes is established to offer the basis to control teaching sequence effectively. A strengthened study strategy based on fault diagnosis is proposed. Test results are analyzed to find students’ knowledge flaws, and according to the knowledge expression model a proper study path needed to be strengthened is deduced to achieve the goal to provide pertinent strengthened teaching for students. According to the students’level of learning, an intelligent algorithm for test paper is proposed. Aiming at the shortcoming of the existing student models, a student model that can reflect cognitive ability is constructed. The method of evaluating cognitive ability, recording study progress and computing mood in study is proposed.With the research on Multi-Agent and J2EE technology, the thesis realized the system based on JADE and J2EE. To some extent this network teaching system realizes the intelligent teaching and provides more convenient and highly effective network-learning environment for learners. Finally, "Java programming" courses as an example of the system provides the results of system experimentation and verifies the effectivity and practicality.更多还原