Self Reported

Self-reported data, encompassing emotional states and experiences, is crucial for understanding human behavior and well-being but presents challenges due to data sparsity and the influence of temporal factors on reporting accuracy. Current research focuses on improving the accuracy of emotion recognition using machine learning models, such as convolutional neural networks and transformer-based approaches, often incorporating multimodal data (e.g., physiological signals, speech) to overcome limitations of solely relying on self-reported data. These advancements hold significant potential for improving the assessment of emotional states in various contexts, including mental health treatment, game design, and chronic disease management, by enabling more accurate and efficient data collection and analysis.