A Sentiment Analysis Model for College Students'Online Learning based on Multimodal Data Fusion
In order to solve the problem of"lack of emotion"of college students,understand the emotional state of college students in online learning,and help teachers to intellectualize teaching and students to personalized learnin,this paper integrates the course comments of college students'online learning platform,facial expressions and gestures during learning,and uses the deep learning method to construct a Bi-LSTMFN sentiment analysis model based on context en-hancement.The model includes four parts,namely,context feature representation,cross-modal information interaction,multi-modal information fusion and emotion recognition.This model can identify whether the emotional state of college students is positive,neutral or negative in online learning,so as to help teachers improve teaching strategies,improve teaching effects and improve students'autonomous learning ability.
NETL
NSTL
万方数据
