Stress Detection Based On Multimodal Data Using Machine Learning Techniques
Stress affects everyone differently but it can lead to a variety of health issues. Early detection of stress can prevent many stress-related health problems. Physiological stress can be identified by basic parameters like heart rate, pulse rate, face recognition, respiratory signals, which provide detailed information about the person state of mind. These parameters vary from person to person on the basis of certain things such as their body condition, age, and gender.
Physiological sensor analytics is becoming an important tool to monitor health. Physiological multi-sensor studies have been conducted previously to detect stress. This paper focuses on features like respiration rate, pulse rate and facial expressions that can now be performed with Microsoft Kinect Xbox 360 sensor, Pulse sensor and Camera, to develop an efficient and robust mechanism for accurate stress identification. Using machine learning algorithms on the above features high accuracy in detecting the stress can be achieved
Index Terms: Physiological stress, Physiological multi-sensor, Microsoft Kinect Xbox 360 sensor, Pulse sensor