EEG-based Stress Detection using DEAP Dataset and Support Vector Machines

A large portion of the population is affected by stress. Early detection is very important to prevent its detrimental effects on the mental and physical health of individuals. One promising technique for detecting stress is electroencephalography, which can monitor the activity of the brain in real time. In the past few years, the use of machine learning algorithms for stress detection has been widely acknowledged. This paper aims at investigating the potential of support vector machines (SVMs) in the DEAP dataset for detecting stress. The paper introduces the concept of stress detection and discusses the use of both electroencephalography (EEG) and SVM in this field. It also reviews the literature on the subject. The methodology section provides an overview of the various steps involved in the development of the study, including the data collection, preprocessing techniques, and the training and evaluation of the SVM model. The paper shows that SVM models can accurately detect stress with the help of EEG data collected from the DEAP database. They performed well with extraction and feature selection techniques, and the combination of PCA and wavelet transform led to the highest accuracy. The discussion section of the paper provides an analysis of the findings, as well as comparisons with previous research. It also explores the potential applications of this technology in the management of stress. The concluding section of the paper summarizes the major findings, limitations, and suggestions for future research. The findings of the study indicate that the use of SVM models in the DEAP database can effectively detect stress, and their potential applications in the field of stress management are immense. The study also emphasizes the importance of optimizing the model's extraction and feature selection procedures.