Architecture for the identification of academic stress levels using Machine Learning and Internet of Things

Abstract

Stress is a mental illness that causes serious health problems and even death when it is not detected and treated in time, although it positively and negatively influences many aspects of people's lives, studies show that it occurs more frequently in the academic life and generates a significant imbalance that occurs through various symptoms, whether physical, psychological or social, some of the most common symptoms are nausea, nervous spasms, depression, difficulty relaxing, lack of concentration among others, which causes that academic performance becomes deficient and negatively impacts the life project of the students, that is why, in this research, the architecture of a software module is proposed that allows identifying the stress level of university students through the use of Machine Learning, specifically Support Vector Machine (SVM) and the Internet of Things paradigm like non-invasive wearable devices.