基于PSO-RF的乡村社区居家养老意愿预测和因素分析——以汕头市潮阳区为例  

Research on Predicting Rural Community Home-based Elderly Care Willingness Based on PSO-RF:A Case Study of Chaoyang District,Shantou City

作　　者：蔡小芳 张楚烽 CAI Xiaofang;ZHANG Chufeng(The Open University of Shantou,Shantou,Guangdong,China,515000;Chongqing Jiaotong University,Chongqing,China,400000)

机构地区：[1]汕头开放大学,广东汕头515000 [2]重庆交通大学,重庆400000

出　　处：《广东开放大学学报》2024年第6期103-111,共9页JOURNAL OF GUANGDONG OPEN UNIVERSITY

基　　金：2024年度汕头市哲学社会科学规划项目“关于乡村养老服务现状调研及提升策略,赋能‘百千万工程’——以汕头市潮阳区为例”(ST24GJ45)。

摘　　要：目前,我国乡村老龄化严重,乡村养老服务体系建设成为社会关注的焦点。社区居家养老方式具有经济性、专业性、互助性等诸多优势,成为缓解农村老龄化困境的新途径。经调查了解汕头市潮阳区乡村老人的生活状况,并对他们是否愿意选择社区居家养老进行问卷调查,对问卷结果采用随机森林(RF)、粒子群算法优化的随机森林模型(PSO-RF)、人工神经网络(ANN)、支持向量机(SVM)四种机器学习模型进行预测,结果显示:PSO-RF模型的准确率和AUC值最高,分别为84.6%和0.83,预测性能优越,具有实际应用价值。把PSO-RF模型应用于乡村社区居家养老意愿预测分析中,所得到的模型有效性及其主要影响因素的结论,能够为乡村智慧养老服务的数据预测提供参考。At present,rural areas in China are facing serious aging problems,and the construction of rural elderly care service systems has become the focus of social attention.The community-based home care model has many advantages such as economy and saving social resources,and has become a new way to alleviate the dilemma of rural aging.This study investigated the living conditions of rural elderly people in Chaoyang District,Shantou City and their willingness to choose community home-based elderly care.Four machine learning models,RF,PSO-RF,ANN,and SVM,were used to predict the willingness of rural community home-based elderly care.The prediction results show that the PSO-RF model has the highest accuracy and AUC value,which are 84.6%and 0.83 respectively.It has superior prediction performance and practical application value.This study provides an effective method for predicting rural community home-based elderly care willingness through smart digital technology,reveals the main factors affecting the willingness to provide elderly care,provides guidance and decision support for accurately promoting new elderly care methods and improving community home-based elderly care services,and is of great significance for solving the people's livelihood issues of rural elderly care services.

关 键 词：社区居家养老 意愿预测 因素分析 机器学习 PSO-RF 

分 类 号：G434[文化科学—教育学] G777[文化科学—教育技术学]