Abstract
This project uses machine learning algorithms to build health status prediction models by analyzing multi-dimensional health behavior data such as exercise frequency, type, sleep duration, and water intake. The authors collected datasets on health indicators and further explored how these lifestyle factors collectively influence health indicator values. The health data collected is used to calculate the target variable “health value,” which is the column that distinguishes whether an individual is healthy or not, based on comparing the mean value of the column, with 1 and 0 indicating healthy or unhealthy. By analyzing the data and machine learning algorithms, they discovered the essential and inseparable relationship between these health indicators and an individual's health status. They optimized the model to improve the accuracy of the prediction. They also confirmed the generalization ability and performance of the model by giving the optimized model new health data and obtaining health values, as well as whether it is healthy. The results of this project highlight the importance of integrating data on lifestyle habits for health prediction to give data support to the development of a scientific basis for personalized health management strategies and health promotion interventions.
Original language | English |
---|---|
Title of host publication | Prediction of Health Indicators Based on Exercise, Sleep, and Water Intake: Data-Based Machine Learning Approach |
Publisher | IGI Global |
Publication status | Published - 2024 |