TESTING THE FACTORS AFFECTING CHILDCARE SERVICE DEMAND UNDER THE MACHINE LEARNING PARADIGM

This study uses the decision tree algorithm in machine learning to explore the impact of childcare service accessibility and perceived value on parents' childcare needs. The study uses two decision tree algorithms, CHAID and CART, to analyze large-scale survey data. The results show that in terms of childcare service accessibility, quality accessibility has the greatest impact on parents' childcare needs; in terms of perceived value of childcare services, the impact of functional value perception is the most significant. The study also compares the impact of different data splitting methods on the model prediction accuracy. The findings of this study provide data support and a decision-making basis for formulating childcare service policies and improving the quality of childcare services.