Nuanced Emotion State Identification: Insights from a CNN-RF Framework

Using facial expressions as a foundation for the classification of emotional states, this study provides a one-ofa-kind CNN-RF model developed specifically for the purpose. An intensive review of precision, recall, F1 score, support, and accuracy metrics over a wide range of emotion classes is used in order to offer a comprehensive evaluation of the efficacy of the model. According to the data, there is a high degree of accuracy, with the Anger class being particularly noteworthy at 91.83%. This draws attention to the fact that the model is able to accurately identify circumstances that are associated with certain feelings that are experienced by the human population. The model’s capacity to properly record genuine occurrences of emotions is shown by solid recall measures, such as the Disgust class, which has an astonishing 87.70% of the total. In light of the fact that the Contempt class achieved an impressive F1 score of 95.19%, the F1 score provides more evidence that the model is indeed efficient in its entirety.