【摘要】 文章采用多Agent技术对中文Web信息进行智能检索,系统结构和工作原理的介绍明确了Agent行为规范,并为一系列信息检索技术提供实现平台。其中,为提高系统效率并实现中文Web信息的智能分类和检索,将支持向量机引入智能过滤子系统。实验结果表明,该系统有效地提高了检索过程中的查全率和查准率。更多还原

【Abstract】 The paper describes some innovations related to using multi-Agents for retrieving Chinese Web information automatically.The introduction of system framework and operating process allows one to describe the Agent’s behaviors easily and to experiment with some key Information Retrieval strategies.In order to improve the system effectiveness and to realize intelligent classifying and retrieving of Chinese Web information, we lead Support Vector Machine (SVM) into intelligent filter subsystem.Experiment results prove a good example of real implementation of Agents for better recall ratio and precision ratio.更多还原