面向在线学习的多模态情感计算研究

Research on Multi-model Affective Computing for Online Learning

The rapid development of computer network technology and multimedia technology has leaded to much attention of educators and learners to the e-learning method that breaks through the limit of time and space to learning activities. But the lack of learners’ affective interaction in the process of e-learning is a big problem in the educational research. To solve this problem， this paper puts forward a multi-modal affective computing model for e-learning based on the existing affective interaction research results and the affective computing theory. The model consists of four modules， namely online learner’s affective data collection， affective data processing and analysis， affective data visualization， and e-learning feedback and intervention. Based on this model， an affective computing system has been developed to achieve affective computing in the e-learning activities based on the learner’s facial expression， speech and text information. The feasibility and effectiveness of the presented system are analyzed through practical test and application.