A hybrid AIoT model for workplace stress prediction and intervention using social media platforms.

Nowadays, stress-related issues are more challenging for employees in the workplaces. As attitudes and practices change in the workplace, employees are more likely to feel stressed. Employees are struggling to maintain their work-life balance. Traditionally, it is very difficult to assess the level of stress accurately, which would have overcome several unwanted incidents. In this study, the investigation of stress in working employees using IoT and machine learning techniques is presented. The proposed machine learning model is used to train the input dataset which is available on five-point-scale questionnaires. These data are available on social media platforms. To forecast results from Internet of Things (IoT) data, an experimental study is conducted using the Naïve Bayes Algorithm (NB), Support Vector Machines (SVM), and Linear Regression (LR). The models' respective accuracy was assessed. Critical analysis conducted through the experiment was used to identify important variables that influence stress. Based on these findings, companies can aim to lower stress levels and give their workers a far more comfortable work atmosphere.