基于协同过滤推荐算法的图书馆智慧服务模式研究

Research on Library Intelligent Service Model Based on Cooperative Filtering Recommendation Algorithm

为了提高图书馆智慧服务能力,进行图书馆智慧服务的推荐算法设计,提出基于协同过滤推荐算法的图书馆智慧服务模式。采用语义抽取方法进行图书馆智慧服务的信息检索,提取图书馆馆藏资源的语义相关性特征量,构建图书馆智慧服务的语义关联规则特征分布集,采用本体特征映射方法进行图书馆智慧服务过程中的文本信息推荐,根据读者的检索偏好进行语义相关性特征配准,建立图书馆智慧服务的读者偏好关联规则数据集,采用模糊聚类方法进行特征分块处理,结合Kalman滤波方法对干扰信息进行协同过滤,实现图书馆智慧服务推荐与读者偏好信息的协同匹配,完成图书馆智慧服务的协同过滤推荐。仿真结果表明,采用该方法进行图书馆智慧服务模式下的协同推荐准确度较高,推荐过程中的时间开销较短,提高了图书馆智慧服务水平。

In order to improve the intelligent service ability of library and design the recommendation algorithm of library intelligent service,a library intelligent service mode based on collaborative filtering recommendation algorithm is put forward.The information retrieval of library wisdom service is carried out by using semantic extraction method,the semantic correlation feature quantity of library collection resources is extracted,and the feature distribution set of semantic association rules of library wisdom service is constructed.The ontology feature mapping method is used to recommend the text information in the process of library intelligence service.According to the user's retrieval preference,the semantic correlation feature registration is carried out,and the user preference association rule data set of library intelligence service is established.The fuzzy clustering method is used for feature block processing,and the Kalman filtering method is combined to filter the interference information cooperatically.The collaborative matching between library wisdom service recommendation and reader preference information is realized,and the collaborative filtering recommendation of library wisdom service is completed.The simulation results show that the cooperative recommendation accuracy under the library wisdom service mode is high,and the time cost in the recommendation process is short,which improves the library wisdom service level.