Health Status Assessment Using Lifestyle Indicators

Dept. of CSE(M. Tech DSE) Ramaiah University of Applied Science Bengaluru, India
Dept. of CSE(M. Tech AIML) Ramaiah University of Applied Science Bengaluru, India
Dept. of CSE Ramaiah University of Applied Science Bengaluru, India
Dept. of CSE Ramaiah University of Applied Science Bengaluru, India

This email address is being protected from spambots. You need JavaScript enabled to view it.
This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Technological development is undergoing rapid proliferation of rapid advancement in technological developments made possible the ever-continuous collection of structured lifestyle and physiological data. Leveraging such data for predictive analytics in health-care is crucial to early detection and analysis of adverse health risks. As a conclusion, we use machine-learning based classification to classify health status classes with respect to different lifestyle elements. We performed a fairly systematic data preprocessing step: outlier removal, categorical encoding, feature normalization, and stratified traintest split. The various classification algorithms were tested and analyzed using these data: Accuracy, Precision, Recall, Fl-score, and Confusion Matrix analysis for Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting (XGBoost) used. By using the experimental results, Logistic Regression was able to achieve the best classification accuracy of 98.66%—a clear outperformance over ensemble-based techniques. This gives a very strong signal that the data set is extremely linearly separable. Linear models tend to do better when they match features of the data. This approach highlights the importance of proper preprocessing and fair testing of analytic health systems and provides a strong, scalable foundation for developing smart preventive health monitoring systems.