Emotion recognition is a fundamental aspect of human communication and understanding. In recent years, advancements in machine learning and deep learning techniques have provided powerful tools for deciphering emotional states from various data sources, including EEG brainwave data. This research paper presents a comprehensive analysis of emotion recognition using a meticulously collected EEG brainwave dataset from Kaggle. The dataset comprises 2132 samples with 2549 features, each labeled with one of three emotions: neutral, negative, and positive, creating a balanced dataset. To assess the dataset’s potential, we implemented ten machine learning algorithms alongside a novel deep learning approach. Our findings revealed that the proposed deep learning algorithm achieved remarkable results with an accuracy, precision, recall, and F1-score all at 0.9977. In comparison, the best-performing machine learning algorithm, “NearestCentroid,” demonstrated high accuracy, precision, recall, and F1-score, all at 0.9953. This study not only underscores the feasibility of recognizing emotions from EEG data but also highlights the efficacy of deep learning techniques in achieving superior performance. This research contributes to the growing body of work on emotion recognition, demonstrating the potential for practical applications in fields such as affective computing, human–computer interaction, and mental health monitoring. The insights gained from this study pave the way for future advancements in emotion recognition technology, ultimately enhancing our understanding of human emotions and their implications across various domains.