Emotion Classification of subjects while watching video, using Recurrent Neural Network and Tensorflow

In the fast-evolving realm of contemporary computing, the challenge of accurately classifying emotions based on physiological signals remains a central focal point. This research embarks on an exploration of the utilization of Re-current Neural Networks (RNN), a potent machine learning technique, for the interpretation of human emotions derived from electroencephalogram (EEG) data. The impetus driving this inquiry stems from the growing demand for non-invasive, real-time emotion recognition systems capable of enhancing human-computer interactions. The EEG data collected from participants who were exposed to emotion-evoking stimuli form the cornerstone of this study. The RNN model undergoes meticulous training and evaluation, leveraging feature extraction from EEG signals. The primary objective is to effectively distinguish between fundamental emotional states, including reactions to positive, negative, and neutral stimuli. Our research outcomes underscore a promising level of accuracy in emotion classification, thus emphasizing the potential of RNN within this specialized domain. Furthermore, these results pave the way for future research directions, such as the integration of multi-modal data and the adoption of more advanced machine learning models to further amplify performance. This study contributes significantly to the field by demonstrating the effectiveness of RNN in EEG-based emotion classification, thereby laying the groundwork for the development of more intuitive and empathetic computing systems.