Sensor networks are becoming increasingly crucial in the realm of technology, as they are present in a diverse array of applications spanning from industrial processes to environmental monitoring, as well as healthcare and wearable devices. The advancement of technologies like wearable devices is presently undergoing notable progress to enhance and render accessible the monitoring of physiological indicators to users. Through the analysis of data collected from these devices and the utilization of contemporary machine learning techniques, it is feasible to attain precise management of the user’s mental and emotional condition in real time. The focus of this study revolves around physiological data and the categorization of emotional states predicated on the gathered physiological data employing multiple machine learning algorithms. A pivotal aspect of the investigation involves the preprocessing stage of the data, encompassing tasks such as data equilibrium and the selection of optimal hyperparameters to heighten classification precision. The analysis of this data has the potential to furnish us with a more profound understanding of diverse matters, such as user health assessment, stress identification, and aiding in physical training.