Research on Influencing Factors of Elderly Care Model Selection for the Silver - Haired Population Based on Traditional Statistical Analysis and Machine Learning Models

Abstract

To address global population aging, this study innovatively integrates statistics and machine learning, analyzing multidimensional data, including economic status, health conditions, family support, and social environment. Logistic regression is employed to elucidate the influencing mechanisms of elderly care model choices, while decision tree and random forest models significantly boost the prediction accuracy to 85%. Key findings indicate that financial capabilities and family support are the core determinants, highlighting machine learning's superiority in identifying crucial influencing factors. This research provides a scientific foundation for optimizing the allocation of elderly care service resources and designing services. Emphasizing the potential of interdisciplinary research, it aims to fundamentally transform gerontology and create a supportive aging future. The developed model can be tailored to specific regional needs, enhancing its applicability and effectiveness, and it also promotes cross - cultural exchanges of elderly care experiences, thus contributing to building a society that enables the elderly to age with dignity and receive adequate support.