基于分类代理的Web信息集成方法的研究  

Research of web information integration based agent of classification

作　　者：李德明[1] 叶进[1] 

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004

出　　处：《计算机工程与设计》2013年第3期1108-1111,共4页Computer Engineering and Design

基　　金：国家自然科学基金项目(61163060)

摘　　要：介绍了现有web信息集成系统的发展现状,分析了各种集成方法的特点及不足,提出了一个通过分类代理方法将现在web信息服务网站进行有效整合集成的新方案。由于引入了语义分类及代理机制,新系统可以最小的代价实现将现有web信息服务网站进行高效的无缝整合,使得用户可并发在多台信息服务器上进行信息检索及发布,从而大大提高用户利用web信息的效率。给出了此设计的软硬件架构组成及原理,详细介绍了此方案的相关特性及核心部件———Agent的分层软件架构、工作流程及系统集成过程等。The development status of the existing web information integration system and the characteristics and shortages of a variety of integration methods are descrobed, then a new method is introduced to implement web information integration sys tem based on classification agent. Because of introduction of the semantic classification and the agency mechanism, the new system can seamless integrates existing web information service at the minimum cost, and can greatly improve the efficiency of the use of web information by allowing users to retrieve and release information to more servers simultaneously. At last, the paper provides not only the system＇ s architecture of the hardware ＆ software, but also more details of the relevant characte ristics of the system, and the core componentsagent＇s layered software architecture, workflow and the integration process of system.

关 键 词：信息集成 代理服务 语义分类 异构数据库 分层模型 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]