Analysis of learning behavior of MOOC enthusiasts has become a posed challenge in the Learning Analytics field, which is especially related to video lecture data, since most learners watch the same online lecture videos. It helps to conduct a comprehensive analysis of such behaviors and explore various learning patterns for learners and predict their performance by MOOC courses video. This paper exploits a temporal sequential classification problem by analyzing video clickstream data and predict learner performance, which is a vital decision-making problem, by addressing their issues and improving the educational process. This paper employs a deep neural network (LSTM) on a set of implicit features extracted from video clickstreams data to predict learners’ weekly performance and enable instructors to set measures for timely intervention. Results show that accuracy rate of the proposed model is 82%–93% throughout course weeks. The proposed LSTM model outperforms baseline ANNs, Super Vector Machine (SVM) and Logistic Regression by an accuracy of 93% in real used courses’ datasets.

由于大多数学习者观看相同的在线讲座视频，因此分析MOOC爱好者的学习行为已成为“学习分析”领域的一个挑战，该领域尤其与视频讲座数据相关。它有助于对此类行为进行全面分析，并为学习者探索各种学习模式，并通过MOOC课程视频预测其表现。本文通过分析视频点击流数据并预测学习者的表现来开发时间顺序分类问题，这是至关重要的决策问题，它可以解决他们的问题并改善教育过程。本文对从视频点击流数据中提取的一组隐式特征采用了深度神经网络（LSTM），以预测学习者的每周表现，并使教师能够制定及时干预的措施。结果表明，提出的模型的准确率在整个课程周中为82％–93％。在实际使用的课程数据集中，所提出的LSTM模型优于基准人工神经网络，超向量机（SVM）和Logistic回归，准确率达到93％。