Non Invasive Human Stress Detection Using Key Stroke Dynamics and Key Stroke Pattern Variations

Abstract

Mental stress causes when we feel we are unable to meet high levels of demands placed upon us. Stress and related illnesses that stress causes has become one of the largest and most costly health problems in the industrialized world. Continuous high levels of stress put a lot of strain on person s psychology and may lead to reduced productivity. Studies have found that long term exposure to high stress levels can give rise to a lot of other illnesses such as depression, high blood pressure and even diabetes. A simple technique which would measure stress levels may significantly help to avoid long-term stress and promote healthy life style. Although it is difficult to measure stress level directly, it is quite possible to annotate stressful events and relate them to physiological signal changes (Ex; Heart Rate variability) that can be easily measured. Most of the existing stress measuring techniques use physiological signal changes to measure stress level. But these techniques are not very suitable for real time stress monitoring. Most of them use body worn sensors or monitoring equipment that needs to be attached to the user s body. This can be very disturbing in his daily routines and normal working patterns. Our research is to propose a non-invasive mechanism which does not demand any additional commitment from the user but still provides an accurate stress level measurement to the user. With the increasing people centricity in contemporary developments of computer science, Affective Computing has become a popular research area. According to the existing research, affective computing has shown positive results in detecting human stress. Our focus in this study is to utilize a readily available yet underutilized resource in Affective Computing, key stroke dynamics (KSD). Recent developments in KSD based affective computing and Biometrics research proves that key stroke variations is a very powerful source of input that provides a valuable insight about an individual's psychological and emotional states. An Increasing number of people use computers in their daily work, responding to emails, accounting, writing reports, programming and etc. Our methodology suggests a personalised approach in detecting stress levels through key stroke variations. An application specific Individual key stroke pattern profile is created for an individual based on his normal typing patterns. This profile consists of trained average values for a set of typing features. Real time stress specific deviations of these features are analysed in order to arrive at the individual stress level