基于机器学习的图书馆用户流失影响因素研讨

论文针对图书馆用户流失问题,基于图书馆用户数据,构建用户特征矩阵,利用深度神经网络预测用户流失风险;将存在流失风险的用户作为重点研究对象,根据用户特征,结合S-O-R理论设计图书馆用户流失影响因素模型。采用扎根理论方法设计问卷,提取影响图书馆用户流失的指标因素,并构建结构方程模型进行分析,将技术研究与质性研究相结合,挖掘影响用户流失的核心因素,并在此基础上,构建了图书馆智慧型服务空间设计模型,为图书馆创新服务提供决策支持。

Research on Factors Affecting Library User Churn Based on Machine Learning

Aiming at the problem of library user churn,based on the user data of the library,the user characteristic matrix is constructed,and the risk of user churn is predicted by using the deep neural network.The users who have the risk of user churn are taken as the key research object,the influencing factors model of library user churn is designed according to the user characteristics and S-O-R theory,and the questionnaire is designed with the grounded theory method to extract the influence.The index factors of library user churn are analyzed by constructing structural equation model.Combining technology research with qualitative research,this paper excavates the core factors affecting user churn,and builds a design model of library intelligent service space on this basis,which provides decision support for library innovative service.