Modeling the relationship between maternal health and infant behavioral characteristics based on machine learning.

Abstract

This study investigates the impact of maternal health on infant development by developing a mathematical model that delineates the relationship between maternal health indicators and infant behavioral characteristics and sleep quality. The main contributions of this study are as follows: (1) The use of Spearman's correlation coefficient to conduct correlation analysis and explore the main factors that influence infant behavioral characteristics based on maternal indicators. (2) The development of a combined model using machine learning techniques, including random forest (RF) and multilayer perceptron (MLP) to establish the relationship between maternal health (physical and psychological health) and infant behavioral characteristics. The model is trained and validated by the real data respectively. (3) The use of the Fuzzy C-means (FCM) dynamic clustering model to classify infant sleep quality. An RF regression model is constructed to predict infant sleep quality using maternal indicators. This study is significant in gaining a deeper understanding of the relationship between maternal health indicators and infant development, and provides a basis for future intervention measures.