加载中…
个人资料
十年磨刀
十年磨刀
  • 博客等级:
  • 博客积分:0
  • 博客访问:2,150
  • 关注人气:0
  • 获赠金笔:0支
  • 赠出金笔:0支
  • 荣誉徽章:
相关博文
推荐博文
正文 字体大小:

[C-E] 基于Agent技术的远程教学平台设计

(2010-10-03 09:07:31)
标签:

杂谈

分类: 汉译英习作

    基于Agent技术的远程教学平台设计

    A Design of Distance Teaching Platform Based on Agent Technology

    摘要:计算机网络技术、卫星通信技术和多媒体技术为远程教育提供了一种新的教学模式。目前现代远程教育系统中存在着通用性不强、灵活性和智能性差等诸多问题。本文通过总结Agent特征和结构,提出了基于Agent机制的远程教学系统平台的设计模型,并给出了具体的实现方法。

    Abstract: The computer network technology, satellite communication technology and multi-media technology offer a new teaching mode for the distance education. Now there are still many problems in modern distance education such as weak generality, flexibility and intelligence, etc. This paper will bring up a design mode of distance teaching platform based on Agent mechanism and the concrete implementation method through analyzing the characteristic and structure of Agent technology.

    关键词:远程教育,Agent技术, 自适应,学生模型

    Key Words: Distance Education, Agent Technology, Self-adaptation, Student Model

    1 前言

    1 Preface

    近年来,随着多媒体技术和网络技术的日臻成熟, 基于Internet/Intranet的远程教学系统得到了前所未有的发展。我国1998 年在高校中推行远程教育,兴起了网上教学研究的高潮,各种网络教学系统纷纷出台,取得了不少成果。 但是,基于Internet/Intranet的远程教学效果和传统课堂教学相比,还有一定的差距,主要存在的问题有:

    In recent years, with increasing maturity of the multi-media and network technology, the distance teaching system based on Internet/Intranet has got the unprecedented development. Since our country started implementing distance education in universities and colleges in 1998, the online teaching research has been very popular, various kinds of teaching systems have appeared on the stage, and many results have been achieved. However, compared with the traditional class teaching, the distance teaching based on Internet/Intranet is still not desirable, with the main existing problems as following:

    (1) 系统缺乏个性化。目前许多的远程教学系统并没有充分考虑到学习者的需求与习惯,所以造成了个性化差、不能提高学生的学习兴趣、教学效果不明显等诸多问题。

    (1) The system lacks individuality. At present, many distance teaching systems haven't made full consideration about learner's requirements and habits, which results in many problems such as poor individuality, failure to improve student's interest, inapparentness of teaching effect, etc.

    (2)缺乏智能化。现代远程教育系统大多是教学材料的堆砌,系统以呈现教学材料为主,学生只是被动地接受教学,且网络信息检索方式缺乏智能化,学生很难从大量的教学资源中检索出自己感兴趣的问题。

    (2) The systems lack intelligence. Most of modern distance education systems just pile up teaching materials, focusing on the presentation of teaching materials, students only accept those teaching materials passively, and the way to search net information lacks intelligence, so it is very hard for students to search something attracting their interest from a great deal of teaching resources.

    (3)系统的动态交互功能不强,学生不能进行发现式、创造式的主动学习,因而很难达到最佳的学习效果。

    (3) The systems don't have strong dynamic interaction, and students cannot have discovery learning or creative learning, so it is hard to attain the best learning effect.

    (4)系统的测试和评价功能不强,缺少考虑学生积极性、参与程序、创造性和学习能力的指标,而且难度级别比较粗略,评价不够准确。

    (4) The system doesn't have strong testing and assessment function, seldom considers the indexes of students' activeness, participating, creativity and learning ability, and is rough in difficulty levels, so it doesn't have enough accurate assessment.

    针对以上基于Web的远程教学系统存在的不足,可以采用Agent技术来提高系统的智能性,使系统可以按照学习者的需求来进行学习,从而提高整个远程教学系统的性能。

    In view of these deficiencies in distance teaching systems based on Web, Agent technology can be used to improve the system's intelligence, makes the system learn according to the learner's requirement, thus improve the performance of the whole distance teaching system.

    2 智能代理技术

    2 Intelligence Agent Technologies

    Agent (智能代理)技术是人工智能和网络技术相结合的产物,是一个具有自适应性和智能性的软件实体,能代表用户或其它程序,以主动服务的方式完成一项工作。也就是说,Agent是一个封装的有独立功能的模块,它包括自己的数据和对这些数据进行操作的算法,它能够接受和处理其他Agent发送来的消息,也能向其他Agent发送消息,是一个有独立解决问题能力、并随环境变化而变化的实体。

    Agent(Intelligence Agent) technology is the outcome of artificial intelligence combined with network technology, which is a software entity with self-adaptation and intelligence on behalf of customers or other programs to accomplish a task by the way of initiative service. That is to say, Agent is an encapsulated module with independent functions, includes its own data and algorithms of operating these data, can accept and process the messages from other Agents and also send messages to other Agents, so it is an entity which has its independent problem-solving ability and can change along with the changing environment.

    Agent包括如下基本特性:

    Agent includes the following basic features:

    (1)自治性:Agent能自行控制状态和行为,能够在没有人或其他程序介入时操作和运行。

    (1) Autonomy: Agent can control its status and behavior by itself and can run without the involvement of anybody or other programs.

    (2)反应性:Agent能够及时地感知和响应其所处环境的变化。

    (2)Reaction: Agent can timely feel and respond to the changes of its surrounding environment.

    (3)能动性:Agent能够主动表现出目标驱动的行为,能够自行选择合适时机采取适宜动作。

    (3)Initiative: Agent can actively perform target-driven behaviors and choose by itself right opportunity to take proper actions.

    (4)可通信性:Agent能够用某种通信方式与其它的实体交换信息和相互作用。

    (4) Communicability: Agent can exchange and interact with other entities in some communication way.

    (5)协作性:Agent之间可以通过通讯语言交流协作。

    (5)Collaboration: Exchange and collaboration can be done between Agents.

    (6)智能性:Agent能够基于学习知识和经验进行相关的推理和智能计算。

    (6) Intelligence: Agent can do related reasoning and intelligent computing based on learning knowledge and experience.

    Agent像是一个黑盒子,其结构模型如图1所示。Agent通过接口感知外部环境,与之进行交互,通过传感器接受外部信息,并对输入信息进行过滤和分类,把信息传递给推理机,推理机根据知识库的知识和规则进行推理并做出决策,把动作指令交给效应器,通过接口对外部环境产生动作。

    Like a black box, Agent's construction model is as figure 1. Agent perceives the external environment and interact with it through the interface; Agent accepts outer information by sensor, filter and sort the input information, then send it to the reasoning machine, which can do reasoning and make decisions according to knowledge and rules in the knowledge base; eventually, hands action instructions to the effectors and produce operations on the external environment through interface.

    3 基于Agent的远程教学平台

    3 Distance Teaching Platform Based on Agent

    近年来,人们对远程教学和教学的智能化进行了广泛的研究。在此基础上,我们设计出一个基于Agent的远程教学系统平台,旨在融合个性化学习和协作学习两种学习观的优点,构建一个可以实现学习的个性化和协同化的远程教学系统,其模型如图2所示。

    In recent years, people have made broad research about the distance teaching and the intelligence of teaching. On this basis, we have designed a distance teaching system platform based on Agent with the purpose of integrating personal learning's and cooperative learning's advantages and building a distance teaching system which can realize personal and cooperative learning. Let's look at its model in Figure 2.

    整个现代远程教学系统包括3个部分:(1)基本数据库,包括:学习资源库、试题库、系统知识库、教学策略库、教学目标库和学生模型库;(2)基本用户,即教师、学生和管理者;(3)Agent,包括:教师Agent、学生Agent、管理者Agent、人机交互界面Agent、教学Agent和管理Agent。

    The whole modern distance teaching system includes three parts: (1) Basic Database, including learning resources library, test questions library, system knowledge library, teaching strategies library, teaching objectives library, and student models library; (2) Basic Users, i.e., teachers, students and administrators;(3) Agent, including teacher Agent, student Agent, administrator Agent, human-machine interface Agent, teaching Agent and administrating Agent.

    下面具体介绍部分Agent的功能。

    Let's introduce some functions of Agent.

    学生Agent:学生登录远程教学系统以后,系统会自动生成一个学生Agent。学生Agent一方面要为对应的学生提供交互界面,引导学生的学习,并在学习过程中根据学生的实际情况,通过教学Agent从教学策略库中选择合适的策略给学生以指导,在学习结束后将学生的学习结果返回学生模型库。同时学生Agent还可以通过与教师Agent进行通信,以实现教学对学生进行辅导、答疑、讨论等。

    Student Agent: After a student logs on the distance teaching system, the system will automatically create a Student Agent. On the one hand, the Student Agent system will offer the student an interactive interface, guide the student to learn, and choose proper strategy from the teaching strategies library through the Teacher Agent to give the student the guidance according to the student's actual condition in the process of learning, then return the student's learning result to the student model library after the student finishes the learning; on the other hand, the Student Agent can tutor the student, answer questions and do discussions by communicating with the Teacher Agent.

    教师Agent:教师登录远程教学系统以后,系统会自动生成一个教师Agent,教师Agent通过与教学Agent交换信息实现作业的发布,考试试题的发布等。还可以与学生Agent交换信息,了解学习者的学习状况和疑难问题,从而动态地、有针对性地对学生实施教学或答疑讨论等。

    Teacher Agent: After a teacher logs on the distance teaching system, the system will automatically create a Teacher Agent. The Teacher Agent can distribute assignments and test questions by communicating with the Teaching Agent. In addition, it can learn the student's learning situation by exchanging information with the Student Agent, thus pertinently implement the teaching, answering questions and discussion, etc.

    管理者Agent:教师登录远程教学系统以后,系统会自动生成一个管理者Agent,管理Agent通过与管理Agent交换信息实现对系统的管理,如:学生管理、教室管理、用户管理、资源的管理和安全的管理等。

    Administrator Agent: After a teacher logs on the distance teaching system, the system will automatically create an Administrator Agent. The Administrator Agent can manage the system by communicating with other Administrator Agents, such as student's management, classroom management, user's management, resources management, security management, etc.

    人机接口Agent:人机接口Agent主要负责学生、教师、管理者与教学、管理子系统之间的交互联系,并记录交互过程。同时,对学生的学习状态、学习进度、学习效果和学习能力进行记录,以便为不同的学生提供个性化教学。

    Human-Machine Interface Agent: It chiefly takes charge of students, teachers, administrators and teaching, managing the relationships between subsystems, and recording the interactive process. Meanwhile, it can record every student's learning status, progress, effect and learning ability in order to offer different students individual teaching.

    教学Agent:是整个教学系统的核心,通过与教师Agent、学生Agent进行通信来为教师和学生提供服务,完成教与学功能。教学Agent完成的主要功能包括:

    Teaching Agent: As the core of the whole teaching system, it offers teachers and students various services and accomplishes functions of teaching and learning by communicating with Teacher Agents and Student Agents. The main functions of Teacher Agent include:

    作业发布:教师利用Agent的学习和推理机制,通过分析学习者对教学内容的浏览、下载等行为确定学习者的兴趣、愿望以及知识水平,从教学资源库中产生因材施教的学习内容,以及从题库中生成有针对性的练习,学生Agent在每次学习之后更新学生模型库,给出下一步学习的建议。

    Assignments Distribution: With the Agent's learning and reasoning mechanism, by analyzing the learner's behaviors such as browsing and downloading the teaching materials, the teacher can confirm the learner's interest; wish and knowledge level, and then generate learning contents in accordance of the student's aptitude from the teaching resources library as well as generating the pertinent exercises from the test questions library. The Student Agent will update the student model library after every learning is finished and offer some suggestions for the next learning.

    网上答疑和问题讨论:学习者在遇到疑难时应该得到多方面的帮助,答疑和问题讨论是基于web的远程教学系统的重要功能,这些功能可通过Agent间的通讯来实现。对于一些具有共性的问题或共同感兴趣的话题,教师Agent可以在各学生Agent能够存取的电子黑板上予以答复或对讨论提出指导性意见。

    Online Answering Questions and Discussing Problems: Learners should get various help when they come across difficulties, The important functions of distance teaching system based on Web include answering questions and discussing problems which can be realized by communication between Agents. To some common topics attracting students' interest, Teacher Agent can answer questions or give guiding opinions for the discussion on the electronic blackboard accessible to each Student Agent.

    远程考试:考试是对学习效果的检验和评测。远程教学系统的考试由教师Agent从学生模型库中分析学生的平均水平,结合学习目标确定试题难度,从题库中抽取、组织试卷进行考试,并及时生成考试成绩以及成绩分析。

    Remote Test: Test examines and evaluates the learning effect. With the analysis of students' average level from the student model library as well as the learning objective, the Teacher Agent determines the difficulty degree of test questions, then extract from the questions library and organizes the test, as well as timely generates the students' test scores and related score analysis.

    远程学习:通过与学生Agent交换信息,实现个性化、智能化的学习,并对用户的学习情况进行记录和分析,给出指导性的建议。

    Remote Learning: It is an individual and intelligent learning by exchanging information with Student Agent, and it can record students' learning conditions and give guiding suggestions.

    管理Agent:管理Agent负责教学过程中的一系列的管理,如课程管理、学籍管理、成绩管理,用户管理等。

    Administrator Agent: It takes charge of a series of management during the teaching process, such as course management, student-status management, score management, user's management, etc.

    4 主要Agent的设计

    4 Main Design of Agent

    Agent定义可描述为一个5元组的实体,即Agent=<A,I,S,T,K>,其中:A为名称或标识,应有惟一性,并可描述Agent的类型;I为界面定义,描述呈现在用户面前的人机界面和通信接口(包括通信协议和I/0接口),一般遵循界面与功能相分离的原则;S为状态集,是描述Agent内部状态的集合,Agent的行为实际上是由一种状态转移到另一种状态的过程;T为事务集,定义Agent的行为,体现Agent的职能;K为知识源,描述Agent行为所需的知识、数据、推理规则以及有关资源,体现为数据结构、数据库(或知识库)等。

    Agent can be defined as a 5-dimensional entity, namely, Agent=< A, I, S, T, K>, among them: A represents name or mark, which should be unique and describes the Agent's type; I represents interface, which describes the human-machine interface and communication interface (including communication protocol and I/O interface), and usually follows the principle of separation between interface and function; S represents status set, which describes the Agent's inner status, actually, Agent's behavior is the process from one status to anther status; T represents transaction set, which defines Agent's behaviors and embodies Agent's functions; K represents knowledge source, which describes the knowledge, data, reasoning rules and related resources required by Agent's behaviors in the form of data structure, database (or knowledge base), etc.

    4.1 教师Agent的设计

    4.1 Design of Teacher Agent

    教师Agent的任务是建立一个教师模型,在系统中扮演教师的角色。教师Agent的设计模型如图3所示。

    Teacher Agent's task is to create a teacher model, playing the role of teacher in the system. Its design model is shown in figure 3.

    1、教师Agent的定义描述如下:

    1, Teacher Agent is described as follows:

    <教师Agent>::=<AgentID,接口,教学控制器,知识库>

    <Teacher Agent>::=<AgentID, interface, teaching controller, knowledge base>

    <教学控制器>::=<教师功能模块,教学推理机>

    <Teaching Controller>::=<Teacher Function Module, Teaching Reasoning Machine>

    <教师功能模块>::=<题库查询、更改、增加、删除,学习资源库查询、更改、增加、删除,讨论与答疑区的启动、管理,远程考试,资源库的查询、更改、增加、删除>

    <Teacher Function Module>::=<Questions Base's Query, Editing, Adding and Deleting; Learning Resources Base's Query, Adding and Deleting; Discussion & Answering Questions Area's Starting and Managing; Remote Test; Resource's Query, Editing, Adding and Deleting>

    <教学推理机>::=<教学策略的选择,知识库的查询、更改、增加、删除>

    <Teaching Reasoning Machine>::=<Choice of Teaching strategies, Knowledge Base's Query, Editing, Adding and Deleting>

    <知识库>::=<学习资源库,教学策略库>

    <Knowledge Base>::=< Learning Resource Library, Teaching Strategy Library>

    2、相关数据库描述如下:

    2. Related Databases are described as follows:

    学习资源库:存放教学内容,可通过接口或控制器更新教学内容。

    Learning Resource Library: It stores the teaching contents, which can be updated through the interface or controller.

    教学策略库:教学策略是对完成特定教学目标而采取的教学活动程序、方法、形式和媒体等因素的总体考虑。它建立在学生认知模型的基础上。

    Teaching Strategy Library: Teaching strategy is a general consideration about the teaching activity procedures, methods, forms and medias which are adopted in order to accomplish the specific teaching objectives. It is based on the student's cognitive model.

    4.2 学生Agent的设计

    4.2 Design of Student Agent

    学生Agent的任务是建立一个学生模型,在系统中扮演学生的角色。学生Agent的设计模型如图4所示。

    Student Agent's task is to create a student's model, playing the role of student in the system. Its design model is shown in figure 4.

    1、学生Agent的定义如下:

    1. Student Agent is described as follows:

    <学生Agent>::=<AgentID,接口,个性学习控制器,学生模型库>

    <Student Agent>::=<AgentID, Interface, Individual Learning Controller, Student's Model Library>

    <个性学习控制器>::=<学生类型判断,注册与登录,学习与测试,学生模型库查询、更改、增加、删除>

    <Individual Learning Controller>::=< Judgment of Student's Type, Registration and Logging on, Learning and Testing, Student's Model Library's Query, Editing, Add and Deleting>

    <学生模型库>::=<学生基本信息库,认知状态库,个性特征库,测试记录库>

    <Student's Model Library>::=< Student's Basic Information Library, Cognitive Status Library, Testing Record Library>

    2、相关数据库如下:

    2. Related Databases are as follows:

    个性特征库:包括关于学生学习能力、学习兴趣、学习特征的所有信息。

    Personality Library: Including all information about students such as learning ability, interest, and characteristic.

    认知状态库:记录每个学生在每一个概念领域里的学习进度及其所掌握的知识程度。

    Cognitive Status Library: Recording every student's learning progress and mastery degree in every realm of concept.

    学生基本信息库:记录每个学生的用户名、姓名、性别、年龄等基本信息。

    Student's Basic Information Library: Recording every student's ID, name, sex, age, etc.

    测试记录库:记录每个学生每次测试的成绩、出错情况、对每一部分的知识掌握情况。

    Testing Record Library: Recording every student's score, errors, mastery condition of every part of knowledge in every test.

    5 结束语

    5 Conclusion

    Agent是人工智能和计算机软件领域中一种新兴的技术,其在远程教育中的应用具有很大的优越性。本文在分析传统的基于Web远程教育不足的基础上探讨了用Agent思想解决远程教育的问题,并设计了一个基于Agent的远程教学系统模型,并在实际的教学应用中得到了很好的教学效果。

    Agent is a rising technology in the field of AI and computer software, which enjoys great superiority in the application of distance education. This paper has discussed about solving problems in distance education by using Agent thought on the basis of analyzing the deficiencies of traditional distance education based on Web, and designed a distance teaching system model based on Agent. This model has obtained very good teaching effect in the practical teaching applications.

    译于2005年11月20日。

0

阅读 收藏 禁止转载 喜欢 打印举报/Report
  

新浪BLOG意见反馈留言板 电话:4000520066 提示音后按1键(按当地市话标准计费) 欢迎批评指正

新浪简介 | About Sina | 广告服务 | 联系我们 | 招聘信息 | 网站律师 | SINA English | 会员注册 | 产品答疑

新浪公司 版权所有