基于轨迹聚类的个性化信息服务策略  

Personalized Information Service Strategy Based on Trajectory Clustering

作　　者：高亚瑞玺 汤珊红[2] 

机构地区：[1]中国国防科技信息中心研究生部,北京100142 [2]中国国防科技信息中心,北京100142

出　　处：《情报理论与实践》2017年第6期87-90,共4页Information Studies:Theory & Application

摘　　要：随着互联网与社交网络等技术的发展,个性化信息服务模式正在不断完善,如何快速获得用户的实时信息需求成为个性化信息服务的关键。目前信息需求获取的方式主要是通过用户主动提交信息、个性化调查以及通过数据挖掘等技术来获取用户的行为模式,这些服务策略可能导致信息滞后,从而无法为用户提供准确的信息服务。文章提出基于轨迹聚类的个性化信息服务策略,实时获取用户访问与检索的轨迹数据,结合轨迹聚类算法建立动态的用户检索模型,快速实现实时个性化信息服务。The development of internet technology and social network has brought the improvement of the personalized information service mode. And the key to become the successful personalized information service is acquiring the real-time data of users＇ information rapidly. Most information service institutions collect users＇ requirements by requesting the user to deliver their personalized information, carrying out personal information survey, or relying on data mining technologies. However, these methods have the problem of information delay, which may fail to provide information service accurately. Therefore, this paper presents a new method for personalized information service based on trajectory clustering. This method obtains the trajectory data of the users timely and builds the dynamic retrieval model through trajectory clustering algorithm, which makes it easier for providing personalized information services timely.

关 键 词：轨迹数据 聚类算法 个性化信息服务 服务策略 

分 类 号：G252[文化科学—图书馆学]