E – Health Monitoring System With Diet And Fitness Recommendation Using ML

Abstract

The increasing prevalence of lifestyle-related disorders such as obesity, diabetes, hypertension, and cardiovascular complications has created a demand for intelligent preventive healthcare systems. Conventional healthcare practices often depend on periodic consultations and delayed interventions, which may not provide continuous guidance for maintaining wellness. This paper presents a Machine Learning based E-Health Monitoring System designed to evaluate user health conditions and generate personalized diet and fitness recommendations using structured health data. The proposed system utilizes user inputs including age, gender, height, weight, body mass index, calorie intake, activity level, and existing medical conditions. Several machine learning algorithms, namely Logistic Regression, Random Forest, K-Nearest Neighbors, and Support Vector Machine, are implemented and compared for health risk classification. Data preprocessing techniques such as normalization, feature encoding, missing value handling, and feature selection are employed to improve model performance. A recommendation engine is integrated to suggest customized meal plans, exercise schedules, hydration reminders, and lifestyle improvements according to predicted health status. The complete system is developed using Python with a Tkinter graphical interface to ensure ease of use. Experimental evaluation demonstrates that the Random Forest model achieved the highest classification accuracy compared with other models. The proposed solution provides an affordable and accessible healthcare support platform without requiring wearable sensors or IoT devices.

DOI: https://doi-ds.org/doilink/04.2026-98164668