摘要
机器学习在网络信息处理中占有重要地位。GHunt是一个采用多项机器学习技术的网络信息智能获取与处理系统。首先,这一系统支持分布式的网络信息并行搜索与内容过滤;其次,采用机器学习技术,包括文本分类、聚类,文本概念抽取,从概念层次理解文本信息;再次,基于概念语义空间有效地统一文本信息管理;最后提供高效的基于概念语义的文本信息检索,以及个性化的专题组织与信息推送服务。文中着重阐述了系统中所用到的机器学习技术。
Machine Learning plays an important role in processing web information.GHunt is an intelligent system based on machine learning for web information acquiring and processing.Firstly,the system supports distributed parallel searching and filtering web information.Secondly,the system can distinguish the domain of the web page and understand the document at the concept level by text classification,clustering and concept extraction based machine learning.Thirdly,the system can manage the documents based on semantic concept space.At last,the system can efficiently provide text retrieve based on semantics and individual recommendation for news event.In the paper the machine learning technology applied in the system are described in detail.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第33期189-191,共3页
Computer Engineering and Applications
基金
国家自然基金(编号:90104021
60173017
60073019)
北京市重点自然科学基金(编号:4011003)资助
关键词
网络信息
机器学习
概念语义空间
分类
聚类
Web Information,machine learning,semantic concept space,classification,clustering