节点文献
基于LSTM的解题模型及其在认知诊断过程中的应用
LSTM based problem-solving model and its application in cognitive diagnosis process
【摘要】 评估学生对教学知识点的掌握情况是计算机辅导教学研究的重要组成部分。目前,关于这方面的研究主要集中在认知诊断领域中。由于学生做题过程中受到各种因素的干扰,因此依靠现有的初级概率模型或规则空间分析难以得到准确有效的分析结果。文章针对该问题提出使用LSTM深度学习网络来实现对典型项目反应模式的拟合,构建典型项目反应模式的解题模型,从而为进一步实现学生认知能力诊断奠定了有效的基础。文章所提认知诊断方案具有简化认知诊断设计,可重复操作性高的特点,对于认知诊断的推广及应用具有重要的意义。
【Abstract】 Evaluating students’ mastery of teaching knowledge points is an important part of computer tutoring teaching research. At present, the research on this aspect mainly focuses on the field of cognitive diagnosis. Due to the interference of various factors in the process of students doing the problem, it is difficult to obtain accurate and effective analysis results by relying on the existing primary probability model or rule space analysis. Aiming at this problem, this paper proposes to use LSTM deep learning network to fit the response mode of typical items and construct the problem solving model of typical item response mode, thus laying an effective foundation for further realizing the diagnosis of students’ cognitive ability. The cognitive diagnosis scheme proposed in this paper is characterized by simplification of cognitive diagnosis design and high repeatability, which is of great significance to the popularization and application of cognitive diagnosis.
- 【文献出处】 无线互联科技 ,Wireless Internet Technology , 编辑部邮箱 ,2023年15期
- 【分类号】TP183;G434
- 【下载频次】6