基于多模态数据融合的大学生在线学习情感分析模型
A Sentiment Analysis Model for College Students'Online Learning based on Multimodal Data Fusion
郁文景 1周金芝1
作者信息
- 1. 亳州学院 电子与信息工程系,安徽 亳州 236800
- 折叠
摘要
为研究大学生"情感缺失"问题,了解大学生在线学习时的情感状态,帮助教师智能化教学和学生个性化学习,文章融合大学生在线学习平台的课程评论、学习时的面部表情和姿态动作,运用深度学习方法,构建基于上下文增强的Bi-LSTMFN情感分析模型.模型包括 4 个部分,即上下文特征表示、跨模态信息交互、多模态信息融合和情感识别.该模型可以识别大学生在线学习时的情感状态,帮助教师改进教学策略,提高教师教学效果和学生自主学习能力.
Abstract
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.
关键词
在线学习/多模态融合/情感分析Key words
online Learning/multimodal fusion/emotional analysis引用本文复制引用
基金项目
安徽省教育厅科学研究项目(SK2021A0750)
亳州学院一般教研项目(2021XJXM057)
安徽省级教研项目(2021xsxxkc182)
出版年
2024
NETL
NSTL
万方数据