Abstract:

As a result of the ongoing COVID-19 outbreak, a number of businesses and institutions have adjusted by reducing headcount or enabling employees to work remotely. Due to the long-term nature of the new normal, many employees have reported heightened levels of psychological stress and tiredness as a result of trying to adapt their personal and professional lives to the new normal. This research looks at how data visualization and machine learning algorithms may be used to make predictions about the stress levels of employees. Using the information, we may create a model to identify whether a worker is under- or over-stressed. Here, we offer experimental results that show how the XGB classifier may be used to boost the predictive power of models. Employee productivity is greatly affected by working hours, workload, age, and job ambiguity, as proven by XGB classifier interpretation. The other factors are negligible in light of what has already been discussed. A decrease in employee voice would naturally come from an increase in working hours, job uncertainty, and workload.

Description:

.