Big Data Analytics-Based Predictive Modeling for Stress Management Using Healthcare System

摘要

This paper presents a big data analytics-based predictive modeling for stress management using healthcare system that analyzes collected sensing data to predict personal health status. We consider the stress management service, which provides daily stress levels by comparing the measuredsensing data with the condition of each decision phase in the stress-level prediction model. The stress-level prediction model can accurately classify huge amounts of personal health data into specific stress levels (normal, high, and very high). To predict stress levels, we use personal healthdata, including four types of attributes (heartbeat, blood pressure, sleeping time, and obesity) as the training dataset. The evaluation results show that our stress-level prediction model obtains over 80% of accuracy, precision, and recall performance on average.