Student score prediction method based on deep neural network

The invention discloses a student score prediction method based on a deep neural network, and the method comprises the steps: building a student score prediction model, combining the model with student visual space attention data, carrying out the feature learning through a deep neural network DNN frame, optimizing the data features through three full-connection layers, and carrying out the prediction of the student scores. And finally, classifying results by using a support vector machine (SVM) to realize an optimal prediction effect. According to the method, the academic scores of the students can be effectively classified and predicted, the advantages of deep learning and machine learning technologies are combined, and an efficient tool is provided for score evaluation in the education field. 本发明公开了一种基于深度神经网络的学生成绩预测方法，该方法建立了学生成绩预测模型，该模型结合了学生视觉空间注意力数据，利用深度神经网络DNN框架进行特征学习，通过三个全连接层优化数据特征，最终利用支持向量机SVM对结果进行分类，实现最佳的预测效果。本发明能有效地对学生的学术成绩进行分类预测，结合了深度学习和机器学习技术的优势，为教育领域的成绩评估提供了一个高效的工具。