Search

With the full application of MOOCs online learning, STEM multidisciplinary and knowledge structures have been achieved, but it has also resulted in a massive number of dropouts, seriously affected the learning sustainability of STEM education concepts, and made it difficult to achieve learning effectiveness. Based on the massive STEM learning behavior instances generated by MOOCs, as well as the entire learning periods, this study considers some key explicit and implicit features associated with learning behavior, and achieves the fusion of convolutional neural network and recurrent neural network through data-driven approaches, incorporates long short-term memory mechanism to develop dropout prediction methods and models. Based on the experimental results, we also discuss the relevant problems of dropouts related to STEM learning behavior, explore the key dropout temporal sequences of the learning process, identify related factors that have key impacts on learning behavior, and deduce intervention measures and early warning suggestions. The entire study can provide effective methods and decisions for researching the STEM learning behavior of MOOCs and has strong research feasibility and urgency. [ABSTRACT FROM AUTHOR]

MOOCs might be an important organization way to realize the online learning process. Online technology and sharing technology enable MOOCs to realize the adaptive scheduling of learning resources, as well as the independent construction of learning sequences. At the same time, it also generates a large number of complex learning behaviors. How to mine and predict the value and laws of MOOCs is a difficult problem in learning analytics applied to MOOCs. This study integrates the context information of the learning process of MOOCs, designs a deep learning model based on multi-entity knowledge graph, in order to predict and track the interest propagation of learning behaviors, and realizes the learning trend guidance mechanism supported by multi features and complex relationships. Through sufficient experiments, the effectiveness and reliability of this model are verified, and based on the analysis results, the best path and implementable scheme for the interest propagation of learning behaviors are constructed. The whole research might provide more references for the study of learning behaviors described by multi-entity knowledge graph of MOOCs. [ABSTRACT FROM AUTHOR]

The temporal sequence of learning behavior is multidimensional and continuous in MOOCs. On the one hand, it supports personalized learning methods, achieves flexible time and space. On the other hand, it also makes MOOCs produce a large number of dropouts and incomplete learning behaviors. Dropout prediction and decision feedback have become an important issue of MOOCs. This study carries out sufficient method design and decision analysis on the dropout trend. Based on a large number of learning behavior instances, we construct a multi behavior type association framework, design dropout prediction model to analyze the temporal sequence of learning behavior, then discuss the corresponding intervention measures, in order to provide adaptive monitoring mechanism for long-term tracking and short-term learning method selection, and enable adaptive decision feedback. the full experiment shows that the designed model might improve the performance of the dropout prediction, which achieves the reliability and feasibility. The whole research can provide key technical solution and decision, which has important theoretical and practical value for dropout research of MOOCs. [ABSTRACT FROM AUTHOR]

Early warning recommendation is crucial for tracking learning behavior and represents a significant issue in interactive learning environments. However, an interactive learning environment-based learning process may not always achieve expected goals, leading to inefficient or ineffective learning behavior and negative emotions. Additionally, many learners fail assessments due to these issues. To address this problem, this study proposes relevant test problems for interpretable early warning recommendations based on massive learning behavior instances and potential relationships. We design an applicable learning analysis model, namely a deep-neural network based on the knowledge graph of learning behavior, and verify its feasibility and reliability through extensive experiments and data analysis. Our results demonstrate that the interactive learning process must match multi-factor analysis at different temporal sequences to determine key temporal sequences or intervals. This is limited by the classification of learning contents and interpretable concepts, which provide effective reference for subsequent learning content with similar concept classes and knowledge structures. Our approach recommends effective learning behavior in appropriate temporal sequences as soon as possible or constructs feasible intervention measures to improve learners' participation. This research deepens and expands early warning by proposing a feasible new method and obtaining key conclusions with vital practical significance. [ABSTRACT FROM AUTHOR]

Interactive learning is a two-way learning method of learners independently by using computer and network technology. In the interactive relationships, interactive learning plays a role for learners to achieve the learning purpose, interactive learning has become an important effect of online learning, but it also has many problems that need to be improved. In particular, how to apply the massive data to achieve data-driven early warning needs further research. This study firstly mines the temporal series of learning behavior features, the corresponding data is collected from one online learning platform of The UK Open University, the massive data obtained from the platform have been desensitized, disclosed and shared, which might ensure the comparison and verification of research results. Secondly, we design a early warning mechanism based on temporal memory enhancement model. Through a large number of data training and testing, this method is useful to the analysis of learning behavior features, and has strong effectiveness and reliability. Thirdly, the decision and intervention mechanism are mined and predicted. The whole work is of great significance to the early warning mechanism of interactive learning process, which has strong theoretical value and practical significance. [ABSTRACT FROM AUTHOR]