网络农业专家系统用户代理机制的研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
Internet上大量、异质、分布、动态的信息造成了“信息过载”。在信息充斥的情况下,如何有效的为用户提供基于Internet的资源发现服务已成为一项重要而迫切的研究课题。搜索引擎部分的解决了信息资源发现的问题,然而其结果却不能令人满意。本文针对农业用户计算机技术比较薄弱的特点,提出了网上农业专家系统用户代理机制。目的是通过学习用户的兴趣,观察用户的行为来建立用户兴趣模型,从而为用户提供个性化信息查询服务。本文首先介绍了Agent和多Agent理论和技术:Agent拥有知识、问题解决方法和问题的有关数据以及内置的控制机制;并且具有个性化的特征,有学习和适应能力。同时,Agent应用程序具有许多良好的特性,如:自主性、社会性、反应性和能动性。Agent技术由面向对象技术发展而来,Agent和对象之间必然存在着联系和区别。论文还介绍了Microsoft的新一代面向对象的C#语言,它既继承了C和C++的强大功能,兼具Visual Basic的简单易用性,它利用.NET框架提供的为计算与通信开发的工具和服务。由于Web服务框架的帮助,网络服务就可以看作是C#的本地对象。C#包含了内嵌的组件模型,支持Web上的标准协议,同时,它支持任何组件转换成可以通过Internet被任意平台上的应用程序调用的XML Web
     Services。通过对个性化服务发展历程的简述,指出本项研究的意义所在。在用户参与的情况下,建立用户兴趣模型。对用户个性特征提取算法进行了改进。根据用户需要和用户的任务模型主动、智能和协作地搜索Internet上的农业专家系统信息源,对搜集到的信息进行分析、过滤。该系统能自适应用户兴趣和信息源的变化,满足用户对网上信息、查询跟踪的需求。有效地解决了现有系统在交互方式、自适应用户兴趣等方面的不足,满足人们在信息浏览时的个性化要求。本论文给出了系统的总体设计思路,系统采用分层设计的思想,将整个系统分为用户层、信息处理层和信息收集层,有利于用户按照自己的兴趣爱好选择个性化界面以及个体Agent优化。并对部分Agent进行详细设计和具体的分析。鉴于Agent理论发展还不是很成熟、研究水平和条件的限制,论文还有很多东西值得探索。
A mass of heterogeneous, distributed, and dynamic information on Internet has resulted in "information overload".With the flood of information, it has become an important research issue to provide users with effective service of resource discovery on Internet.Search engines attempt to solve this problem, yet the effect are far from satisfying. The agent mechanism of users about agricultural expert system on network was put forward according to the character that computer technology of agricultural users is weak in this dissertation. Its purpose is to construct the interesting model of a user by learning the interest of the user and observing his(her) behavior so that the service of individualized information inquiry will be offered. Firstly, the theories and technologies of Agent and multi-Agent have been introduced that Agent holds knowledge, methods of solving problems, relative data of problems, built-in controlling mechanism, individualized characteristics, study and adaptive capacity etc. Meanwhile, the application of Agent has many good characteristics such as self-determination, sociality, activity and reactivity etc. Agent technology derives from object-oriented technology, so that relation and difference must inevitably exist between of them. The new object-oriented language of Microsoft company--C# has been also presented. It not only inherits strong functions of C and C+ +, but also keep on being used briefly and easily as Visual Basic. It applies the tools and services that the frame of .NET provides for calculation and communication. With the help of the frame of Web services, network services may be looked on as the native objects of C#. C# includes the embedded pack model, supports the standard protocol of Web and at the same time it also supports XML Web Services which are transformed by any pack and can be transferred by the application of any platform through Internet. The meaning of this research is indicated by the general information of developed course of the personalized services. The interesting model of users is set up under the circumstance of users' participation. The algorithm of distilling personalized characters of users is ameliorated. The information source of agricultural Expert System on internet is searched initiatively, intelligently and cooperatively according to the need and task
    
    
    
    model of users. The system can adapt the interest of users and the change of information source in its own way and satisfy users' demand of searching and tracking the information on the networks. Compared with current system, it has the features of adjusting to shifting user's interests and changing source adaptively, as well as personalized searching results. Integral designment of the system is provided in this paper. The system is separated into user Agent layout, information treatment Agent layout and information collection Agent layout, which follows the idea of delamination. The designment is propitious to the users to choose personal interface according to their interest, it also optimizes the sole Agent. Some Agents are designed and analyzed detailedly. Considering that agent theory is not very well-rounded, and also our level and condition is limited,many of the thesis is needed to study further.
引文
[1] 邢志宇.集成搜索引擎与元搜索引擎[J/OL].http://www.sowang.com/sousuo/20031005.htm
    [2] 汪晓岩,胡庆生,李斌等.面向Internet的个性化智能信息检索[J].计算机研究与发展,1999,39(6):1039~1046.
    [3] Thorstoen Joachims etal. Web Watcher:A Tour Guide for the WWW.http://www.cs.cmu.edu.
    [4] Krulwich B, Burkoy C.The InfoFinder Agent:Learning user interests through heuristic phrase extraction.IEEE Expert, 1997.12(5),22~27.
    [5] 石纯一,张伟等.Agent研究进展[J/OL].http://www.swarmagents.com/thesis/detail.asp?id=108.
    [6] Weiss G(ed). Multi-agent systems, A modern approach to distributed artificial intelligence. Cam-bridge, MA: The MIT Press, 1999.
    [7] 王红卫.建模与仿真[M].北京:科学出版社,2003.
    [8] 史忠植.智能主体及其应用[M].北京:科学出版社,2000.8.
    [9] Stan Franklin, Art Graesser.Is it an Agent, or just a Program?:A Taxonomy for Autonomous Agents, Institute for Intelligent Systems University of Memphis
    [10] 比尔.盖茨.未来之路[M].北京:北京大学出版社,1995.
    [11] Michael Wooldridge, Nicholas R.Jennings.Intelligent Agent:Theory and Practice.The Knowledge Engineering Review, 1995,10(2): 115~152.
    [12] 肖文鹏.移动Agent:智能化分布式计算新模型.http://www-900.ibm.com/developerWorks/cn.
    [13] Michael Wooldridge, Nicholas R.Jennings.Intelligent.Agent:Theory and Practice.The Knowledge Engineering Review, 1995,10(2): 115~152.
    [14] 刘海龙.动态环境下分布式智能系统的任务协作理论研究[D].浙江大学博士学位论文,2001,6.
    [15] 海涛.基于Agent的电子邮件系统的设计与实现[D].昆明理工大学硕士学位论文,2002.3.
    [16] 刘大有,杨鲲,陈建中.Agent研究现状与发展趋势[J].软件学报,2000,11(3):315~321.
    [17] 刘大有,杨博,杨鲲.移动Agent技术的发展、挑战与趋势[J/OL].http://www.swarmagents.com/thesis/detail.asp?id=109.
    [18] Danny B.L.Seven Good Reasons for Mobile Agent[J].COMMUNICATIONS OF THE ACM, 1999,42(3):88~89.
    [19] Gunter.Karjoth.A Sesurity Model For Aglet[J].IEEE INTERNET COMPUTING. http://www.computer.org/internet/.
    
    
    [20] David Kotz. AGENT TCL: Targeting the Needs of Mobile Compmer[J].IEEE INTERNET COMPUTING. http://www.computer.or/internet/.
    [21] 胡朝晖,陈奇,俞瑞钊.移动Agent系统综述[J].计算机应用研究,2000,10:1~3.
    [22] 范玉顺,曹军威.多代理系统:理论、方法与应用[M].北京:清华大学出版社,2002:4~11.
    [23] 刘金琨.复杂工业过程智能控制研究[R].浙江大学博士后研究工作报告,1998,12.
    [24] Randall Davis, Reid G.Smith.Negotiation as a metaphor for distributed problem-solving. Artificial Intelligence, 1983,20.63~109.
    [25] Michael Wooldridge.An Introduction to MultiAgent Systems[M].北京:电子工业出版社,2003,74.
    [26] 田国英.移动Agent体系结构及迁移实现[J/OL].http://www.yesky.com/20010216/159459.shtml.
    [27] 智能科学研究现状[EB/OL].http://www.intsci.ac.cn/mise/research.html.
    [28] About:Agentcities[EB/OL]. http://www.agentcities.org/.
    [29] 王珑,章毅,吴跃.MAS环境下实现Agnet交互协作的关键技术[J].电子科学大学学报,2003,32(2):158~163.
    [30] 魏长华.Agent与面向Agent的程序设计[J].华中师范大学学报,1998,32(3):284~289.
    [31] 尹章才.一种新的编程方法——面向Agent的程序设计[J].信息技术,2002,3:44~45.
    [32] 张士杰,郭宏伟.一个面向Agent的程序设计框架系统[J].计算机工程与应用,2002,12:132~134.
    [33] 赵克立等.ASRNET编程培训教程[M].北京:清华大学出版社,2002:13~22.
    [34] 郑宇军,崔伟宁,王旦伟.Visual C#案例教程[M].北京:北京希望电子出版社,2002:13~15.
    [35] 柴晓路,梁宇奇.Web Services技术、架构和应用[M].北京:电子工业出版社,2003:142~154.
    [36] Timothy Budd.Understanding Object-Oriented Programming With JAVA[M]. 北京:清华大学出版社.2002,13~37.
    [37] 李建忠.C#与Java述评[J/OL].http://www.csdn.net/news/newstopic/4/4697.shtml.
    [38] 周拥峰,焦桢,夏宽理.XML在多代理系统中的应用[J].计算机工程与应用,2002,21:93~94.
    [39] Lu Zhengding, Li Chunlin, Li Layuan.Coordinating Mobile Agents by the XML-Based Tuple Space[J]. Journal of Computer Science and Technology, 2002,17(6):888~900.
    [40] 罗清磊,李卫华.利用Java实现基于XML的Web上的数据库挖掘Agent[J].MODERN COMPUTER,2002,1:13~16.
    [41] Extensible Markup Language(XML)[EB/OL].http://www.w3.org/XML/.
    [42] 杨林,茅玉蓉.个性化:定制你的网络服务[J/OL].http://tech.tom.com.
    [43] Web Watcher:A TourGuide for the WWW[EB/OL]. http://www.cs.cmu.edu/.
    [44] 应晓敏,窦文华.条条道路通罗马——Internet个性化服务的主要形式[N].计算机世
    
    界报.http://www2.ccw.com.cn/03/03221b/0322b51_3.asp.
    [45] 冯翱,刘斌,卢增祥等.Open Bookmark——基于Agent的信息过滤系统[J].清华大学学报(自然科学版),2001,41(3):85~88.
    [46] 潘金贵,胡学联,李俊等.一个个性化的信息搜集Agent的设计与实现[J].软件学报,2001,7.
    [47] 周文辉,张灵玲,李俊.基于Internet的DOL教材信息发现和收集Agent[J].南京大学学报(自然科学),1999,35(1):74~81.
    [48] 张剑,郭燕慧,钟义信.基于特征项的个性化推荐服务模型(C).中国人工智能进展,2003:1101~1105.
    [49] 吕印.基于WWW的个性化信息跟踪检测Agent设计[D].大连:大连理工大学,2000.
    [50] 林鸿飞,杨元生.用户兴趣模型的表示和更新机制[J].计算机研究与发展,2002,39(7):843~847.
    [51] 李卫华.个性化网络信息过滤Agent的反馈评价机制[J].计算机工程与应用,2002,3:158~160.
    [52] 白丽君,张永奎,李荣陆.基于智能Agent的用户兴趣发现和更新[J].计算机工程,2003,29(2):236~237.
    [53] 李小将,刘争元.基于用户兴趣的Web信息检索智能代理[J].指挥技术学院学报,2000,11(5):79~84.
    [54] 徐德智,吴敏,陆文彦.基于Agent的专业搜索引擎的研究和构造[J].计算机工程,2002,28(10):99~101.
    [55] 李进华,焦玉英.基于智能代理的网上专题信息服务模型及算法研究[J].信息系统,2002,25(1):51~54.
    [56] Booch G, Jacobson I,Rumbaugh J.Version 1.0 of the unified modeling language.http://www.rational.com/uml/documentstional.html.
    [57] 王志军,于超.基于隐式反馈的个人信息检索技术及实现[J].计算机工程,2003,29(6):158~159.
    [58] Maes P, Kozierok R.Learning interface agents [J].Proceedings of the 11th National Conference on Artificial Intelligence, 1994:459~465.
    [59] 陈鹏,苏洁.基于Agent的信息检索系统设计与个性化算法实现(C).中国人工智能进展,2003:1034~1038.
    [60] 赵水森.基于因特网的个性化信息服务研究[J].中国图书馆学报,2003,4:20~24.
    [61] 王晓凯,侯朝桢.基于多Agent的网络管理技术研究[J].计算机工程与应用,2002,12:15~17.
    [62] 刘建勋,张申生,王英林.基于Agent的信息系统模糊接口研究[J].情报学报,2001,20(4):464~470.
    [63] 王继成,邹涛,杨小江等.基于Internet的信息资源发现技术与实现[J].计算机研究与发展,1999,36(11):1396~1374.
    [64] 刘树勋.Internet智能搜索Agent研究与实现[D].广东工业大学,2000.
    
    
    [65] 王永庆.人工智能原理与方法[M].西安:西安交通大学出版社,1998.257~294.
    [66] 林尧瑞,马少平.人工智能导论.北京:清华大学出版社,2001,14~98.
    [67] 黄镝.基于信息Agent技术的个性化学术信息系统[J].情报科学,2003,21(4):404~405.
    [68] 陈红英,李卫华,茅革非.智能过滤Agent在Internet上的应用[J].计算机工程,2002,28(3):71~72.
    [69] 李俊,张灵玲,周文辉等.一个智能用户接口Agent设计与实现[J].软件学报,1999.10(8):835~842.
    [70] Pattie Maes, Robyn Kozierok.Learning Interface Agent.Washington,DC:MIT Press, 1993,7.
    [71] 石晶,龚震宇,裘杭萍等.基于用户兴趣模型的智能信息检索系统技术与实现[J].情报学报,2003,22(3):282~286.
    [72] 李燕,樊玮.基于Agent的用户界面模型研究[J]。计算机工程与应用,2002,9:52~54.
    [73] 谢伟凯,蒋长浩,史元春等.“智能教室”——一种用于远程教育的交互空间[J].
    [74] 郭少友.基于XML的分布式信息检索[J].现代图书情报技术,2003,2:51~53.
    [75] 李春兰,刘金德.网络智能信息服务系统的设计与实现[J].信息与控制,2001,30(1):93~96.
    [76] 李桢,倪天倪.基于Agent的智能决策支持系统模型的研究及应用[J].计算机工程,2002,28(5):120~122.
    [77] 蔡自兴,徐光祜.人工智能及其应用[M].北京:清华大学出版社,2000:17~58.
    [78] R.Koblick.Concordia[J].Communications of ACM,1999,42(3):96~99.
    [79] 陈俊杰,薛云,宋翰涛等.基于Agent的元搜索引擎的研究与设计[J].计算机工程与应用,2003,10:33~36.
    [80] 韩立新,黄丽雯,恽爽等.基于Agent的面向Internet的信息检索系统的设计和实现[J].情报学报,2002,21(3):259~263.
    [81] 李晓明,刘建国.搜索引擎技术及趋势[EB/OL].http://www.google8.net/archives/000013.html.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700