基于多Agent技术的分布式信息抽取系统研究  

Distributed Information Extraction System Research Based on Multi-Agent Technology

作　　者：魏保子[1] 王儒敬[1] 

机构地区：[1]中国科学院合肥智能机械研究所

出　　处：《微电子学与计算机》2008年第6期18-21,共4页Microelectronics & Computer

基　　金：国家“八六三”计划项目(2006AA102237);国家科技支撑计划项目(2006BAD10A1410)

摘　　要：讨论了信息抽取的必要性及其现状,并提出一个基于多Agent技术的分布式信息抽取系统模型.系统主要有信息抽取Agent、数据清洗Agent、数据保存Agent等以及相应的知识库组成.并采用分而治之的思想,把信息抽取中遇到的问题分解,分配到各个Agent去完成.提出一种新的规则表示方法,抽取规则可以根据网页结构进行调整,该系统具有一定的自适应性.The information extraction necessity and the present situation is discussed, and a distributed information extraction system model based on the multi-agent technology is proposed. This system is composed of the user interface agent, the agent manager, the information extraction agent, the data clean agent; the data preserves agent and corresponding knowledge library. Using thought called divide and conquer, the question which was met in process of information extraction is decomposed and assigned to each agent to solve. At the same time, the extraction rules are able to adjust themselves to the homepage structure; this system has certain adaptability.

关 键 词：多AGENT技术 信息抽取 分布式 抽取规则 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]