摘要
传统的信息检索方法无法凭借浏览特征行为或信息为用户提供有效的检索目标,导致查准率、检索精度以及稳定性较低,于是提出基于LDAP的大数据浏览隐式反馈信息检索仿真方法。通过LDAP目录服务架构内容,获取用户浏览行为特征信息,明确用户对某种文档的感兴趣程度,以构建用户行为特征模型。用户行为特征模型利用InfoAgent系统来实现,将元搜索引擎与Agent技术相结合,获取到最精准的用户浏览行为,使模型最大限度地反映用户的浏览习惯,最终使检索目标更接近于用户需求。仿真结果表明,所提方法具有较高的查准率,检索精度且稳定性较高。
Traditional information retrieval methods can not provide users with effective retrieval targets by virtue of browsing characteristic behavior or information, resulting in low precision, retrieval accuracy and stability. Therefore, a simulation method of implicit feedback information retrieval for big data browsing based on LDAP is proposed.According to the LDAP directory service architecture, the user browsing behavior characteristic information was obtained. The user’s interest in the document was determined for using the info Agent system to establish the user behavior feature model. Meta search engine and agent technology were combined to obtain the most accurate user browsing behavior. Because the model can reflect users’ browsing habits to the greatest extent, the retrieval goal was closer to users’ needs. The simulation results show that this method has high precision, retrieval accuracy and stability.
作者
叶承斌
李宏亨
YE Cheng-bin;LI Hong-heng(School of Information and Management,Guangxi Medical University,Nanning Guangxi 530021,China)
出处
《计算机仿真》
北大核心
2021年第12期449-453,共5页
Computer Simulation
关键词
隐式反馈信息
轻型目录访问协议
元搜索引擎
用户行为特征模型
Implicit feedback information
Lightweight directory access protocol
Meta search engine
User interest model
