A Learning Emotional State Recognition Method Based on Multi-source Information Fusion

Abstract

The invention discloses a learning emotional state recognition method based on multi-source information fusion. The modified method includes: collecting learners' facial expression images and interactive behavior data from an online learning platform, and obtaining a series of expression images for a certain period of time according to an expression recognition algorithm , feature extraction and classification; extract useful interactive behavior data from Web log files, and use subjective Bayesian methods for uncertainty reasoning and in-depth analysis of data. On the basis of judging the learner's attention to the current learning content, combined with the expression recognition results, the online learning emotional state of the learner is judged, and is divided into emotional states such as "satisfaction" and "confusion". The present invention is applicable to subject teaching of blended learning, and can grasp learners' online learning emotional state in real time through expression recognition, thereby providing more targeted manual intervention. According to the results of emotion recognition, learners' mastery and interest in the current learning content can also be judged.

Description

technical field [0001] The invention relates to the field of intelligent human-computer interaction, in particular to a learning emotional state recognition method based on multi-source information fusion. Background technique [0002] Emotion is an important part of human intelligence, and the correct recognition of human emotion will help the computer to better grasp the human psychological state and serve human beings better. Facial expression recognition, as a research direction of affective computing, constitutes the basis of emotion understanding and the premise of realizing human-computer intelligent interaction. The extreme refinement of facial expressions consumes a lot of computing time and affects the timeliness and experience of human-computer interaction. Therefore, facial expression feature extraction has become an important research topic for facial expression recognition. [0003] Online learning provides learners with a brand-new learning method, breaking t...

AI Technical Summary

Problems solved by technology

However, this form of pedagogy that separates time and space widens the distance between people, resulting in a lack of interaction between teachers and learners, and the phenomenon of "emotional loss" is serious.

Images