【摘要】 针对创新型人才培养与课堂教学质量之间的紧密联系,在对面部表情与课堂教学效果进行关联分析的基础上,利用深度学习技术,重点构建出切实可行的课堂教学效果评价模型并应用于教学实践.借助神经网络学习软件TensorFlow,对日本女性表情数据集模型进行了训练和测试,并将ResNet50算法和Random Forest等4种机器学习算法进行对比分析.实验结果表明ResNet50算法在训练集和测试集中的效果最优,在ROC曲线图中,表现效果也最好,能够有效区别Jaffe数据库中的7种不同表情.本文提出的模型能够获取课堂教学中学生面部表情的变化,从而促进课堂教学质量的提高,由此提供了一种课堂教学质量评价的新方法.更多还原

【Abstract】 In view of the close relationship between the cultivation of innovative talents and the quality of classroom teaching, on the basis of analyzing the relationship between facial expression and classroom teaching effect, using deep learning technology, constructing a feasible evaluation model of classroom teaching effect and applying it to teaching practice were studied in this paper. The model of Japanese female facial expression dataset was trained and tested with the help of neural network learning software TensorFlow, and four machine learning algorithms such as ResNet50 algorithm and Random Forest were compared and analyzed. The experimental results showed that the effect of ResNet50 algorithm was the best in training set and test set, and it was also the best in ROC curve, which can effectively distinguish 7 different expressions in Jaffe database. The model proposed in this paper can obtain the changes of students’ facial expressions in online classroom, promote the improvement of classroom teaching quality, and provide a new method for classroom teaching quality evaluation.更多还原