Deep Regression hybridized Neural Network in human stress detection

In today’s environment, psychological problems such as stress, sadness, and anxiety become significant. Millions of individuals suffer from depression every year, but only a least percentage of them undergo timely treatment. As a result, it becomes critical to identify individual tension and means to relax instantly. It is necessary to recognize and handle stress before it becomes a severe problem. The primary goal is to use Machine Learning techniques to detect anxiety in people to improve their quality of life. The vast majority of stress datasets have been collected while consumers were subjected to high-stress situations, such as music, video files, significant equipment failure, visual information, and gameplay situations. However, whether certain databases are used to construct models that can accurately identify anxiety in various situations is still an open question. This paper offers a hybrid strategy DR-CNN for multimodal stress detection that uses a convolutional neural network (CNN) and a regression classifier. Our findings show that this technique has promise, but they also highlight the need for more research to understand better the relevance of deep-learning approaches in designing generally applicable designs for multidimensional emotion detection.