Pengenalan Emosi Berbasis EEG Tinjauan Terbaru dari Dataset dan Metode

Affective computing and HCI have focused on EEG-based emotion recognition. This paper analyses EEG emotion recognition datasets and methods. This review investigates EEG technology's ability to capture emotional states, revealing people's affective experiences. A comprehensive literature assessment revealed relevant studies and a wide range of datasets and methodologies used in this sector. This analysis investigates DEAP and SEED datasets, focusing on crucial factors including subject number, emotion range, and experimental setting. The review also evaluates each dataset's strengths and weaknesses, focusing on how they affect research results. Filtering, artifact reduction, and time-frequency analysis are used to enhance unprocessed EEG data. This study examines EEG signal characteristics extraction methods for meaningful representations. Statistical, spectral, and time-frequency features are used. In emotion categorization problems, machine learning methods like SVM, k-NN, and deep learning are examined. The review shows how different tactics affect emotion recognition accuracy. This study explores EEG-based emotion recognition in healthcare, entertainment, and HCI. The report concludes by highlighting multimodal fusion, customized emotion recognition, and real-time processing as future research fields. These domains improve EEG-based emotion identification systems' precision, reliability, and application. This evaluation provides scholars and professionals with valuable perspectives to advance EEG-based emotion identification research and practice.