Exploring classroom emotion with cloud-based facial recognizer in the Chinese beginning class: A preliminary study

Abstract

The Facial Emotion Recognizer (FER) detection technology in the education field is in the early stage. Objective, to investigate the classroom emotion and the effectiveness of the Microsoft cloud-based FER interpretations The randomly selected 29 international students who enrolled the fundamental Chinese language course were investigated during five study weeks using the paper-based Student Outcome Survey which measured teaching, assessment, generic skill and learning experience. For FER, the 784 students' facial images were extracted and uploaded into the cloud-based APIs. The students with GPA ≥ 80% showed higher significantly in Assessment components. No significant difference in FER emotion detection was found among five study weeks; however, natural and happiness emotion scores showed symmetric significantly in the opposite direction with r = -0.82, P < 0.1. The students' overall outcomes were over average. FER indicated that the neutral emotion was the highest detected score. The APIs Reliability was still considerably questioned. Using updated quality and technology of photo and video recording were recommended in the further research.