Analysis of Accuracy Between SVM and CNN Algorithms in Facial Expression Recognition

Accurate facial expression recognition (FER) systems are critical for many applications, including human- computer interaction, emotion analysis, and healthcare. This study uses the CK+ dataset, which contains 593 sequences of facial expressions across seven emotion categories, to extensively assess the performance of Support Vector Machines (SVM) and Convolutional Neural Networks (CNN) in Facial Expression Recognition (FER). CNN used a deep learning architecture, whereas SVM relied on face landmark coordinates. These models were evaluated using the criteria of accuracy, precision, recall, and F1 score. SVM accomplished an overall score of 100%, coupled with perfect precision, recall, and F1 scores. CNN also showed impressive results in accuracy and robustness. From the above research studies, it can be observed that both SVM and CNN have gained many successes in FER and thus have been selected according to the characteristics of the datasets and requirements of the applications to provide information beneficial for developers and researchers in the FER system field.