STRESS DETECTION IN IT PROFESSIONALS USING DEEP LEARNING

Abstract

Stress management systems play a significant role to 
detect the stress levels whichdisrupts our socio- 
economic lifestyle. As World Health Organization 
(WHO) says,Stress is a mental health problem 
affecting the life of one in four citizens. Humanstress 
leads to mental as well as socio-fiscal problems, lack 
of clarity in work, poorworking relationship, 
depression and finally commitment of suicide in 
severecases. This demands counselling to beprovided 
for the stressed individual's copeup against stress. 
Stress avoidance is impossible but preventive actions 
helps toOvercoming stress. Currently, only medical 
and physiological experts can determine whether one 
is in a depressed state (stressed) or not. One of the 
traditional method to detect stress is based on 
questionnaires. This methodcompletely depends on 
the answers given by the individuals, people will 
betremulous to say whether they are stressed or 
normal. Automatic detection of stressminimizes the 
risk health issues and improves the welfare of society. 
Thispaves the way for the necessity of a scientific tool, 
which uses physiologicalsignals thereby automating 
the detection of stress levels in individuals. 
Stressdetection is discussed in variousliteratures as it 
is a significant societalcontribution that enhances the 
lifestyle of individuals. Ghaderi et al. analysed 
stressusing Respiration, Heart rate (HR), facial 
electromyography (EMG), Galvanic skinresponse 
(GSR) foot and GSR hand data with a conclusion that, 
features pertainingrespiration processes are substantial 
in stress detection. Maria Viqueira et aldescribes 
mental stress prediction using a standalone stress 
sensing hardware byinterfacing GSR as the only 
physiological sensor.