基于聚类与GRNN算法的图书馆读者借阅行为分析  

Analysis of library readers’ loan behavior based on clustering and GRNN algorithms

作　　者：宋爱香 胡西民[1] 李丹莎[1] SONG Ai-xiang;HU Xi-min;LI Dan-sha(Network&Information Management Office,Xi’an Polytechnic University,Xi’an 710048,China)

机构地区：[1]西安工程大学网络与信息化处,陕西西安710048

出　　处：《电子设计工程》2020年第11期1-4,共4页Electronic Design Engineering

基　　金：国家自然科学基金青年项目(11705135);陕西省教育信息化发展研究项目(18JX014Y)。

摘　　要：通过数据挖掘技术和智能算法寻找图书馆读者的借阅喜好,提高图书馆的管理效率和服务水平。以西安工程大学图书馆读者借阅记录为数据来源,选择读者借阅图书类目和读者所属学院作为参考变量,进行聚类分析,各聚类群组具有相似的借阅模式和显著的借阅特征。利用聚类群组中读者的主要借阅图书类目,建立广义回归神经网络(GRNN)模型,模型诊断准确率为94%,因此,将GRNN模型应用到读者借阅行为分析中。据此指导图书采购、完善图书馆藏、优化个性化推荐服务,为建设智能化图书馆奠定基础。Through data mining technology and intelligent algorithm to find the library readers’ borrowing preferences,improve the management efficiency and service level of the library. Taking the borrowing records of readers in the library of Xi’an University of Engineering as the data source,this paper chooses the categories of books borrowed by readers and the colleges to which readers belong as the reference variables,and carries out cluster analysis. The clustering groups have similar borrowing patterns and significant borrowing characteristics. The Generalized Regression Neural Network(GRNN) model is established by using the main categories of borrowing books of the readers in the clustering group. The diagnostic accuracy of the model is 94%. Therefore,the GRNN model is applied to the analysis of the borrowing behavior of the readers. Based on this,we can guide book purchasing,improve library collection,optimize personalized recommendation service,and lay a foundation for building an intelli gent library.

关 键 词：高校图书馆 读者借阅行为 聚类分析 GRNN 

分 类 号：G250.76[文化科学—图书馆学]