Student performance prediction has attracted increasing attention in the field of educational data mining, or more broadly, intelligent education or “AI + education”. Accurate performance prediction plays a significant role in solving the problem of a student dropping out, promoting personalized learning and improving teaching efficiency, etc. Traditional student performance prediction methods usually ignore the potential (underlying) relationship among students. In this paper, we use graph structure to reflect the students’ relationships and propose a novel pipeline for student performance prediction based on newly-developed multi-topology graph neural networks (termed MTGNN). In particular, we propose various ways for graph construction based on similarity learning using different distance metrics. Based on the multiple graphs of different topologies, we design an MTGNN module, as a key module in the pipeline, to deal with the semi-supervised node classification problem where each node represents a student (and the node label is the student’s performance, e.g., Pass/Fail/Withdrawal). An attention-based method is developed to produce the unified graph representation in MTGNN. The effectiveness of the proposed pipeline is verified in a case study, where a real-world educational dataset and several existing approaches are used for performance comparison. The experiment results show that, compared with some traditional machine learning methods and the vanilla graph convolutional network with only a single graph topology, our proposed pipeline works effectively and favorably in student performance prediction.

The goal of this study is to explore the affordances of spherical video-based virtual reality (SVVR)-based learning activities by comparing the students’ learning engagement models and their writing learning outcomes in a double-loop SVVR-based (DL-SVVR) learning activity and a single-loop SVVR-based (SL-SVVR) learning activity, respectively. For this purpose, we conduct an empirical study involving 82 fourth-grade students. The statistical results show that the students in the DL-SVVR group had higher learning behavioral engagement and better writing learning outcomes than those in the SL-SVVR group. Furthermore, as for the interrelationship between the subscales of students’ learning engagement, we find that, for the students in SL-SVVR group, their behavioral engagement can be positively predicted by the emotional engagement, while in the DL-SVVR group, students’ behavioral engagement can be positively predicted by their social engagement. For both groups, their social engagement can positively predict the emotional engagement. Importantly, our empirical results also show that the double-loop learning approach can potentially promote students’ writing learning outcomes by shifting their social engagement and behavioral engagement. The qualitative analysis results indicate that peer interaction has a positive impact on improving students’ learning interest and writing outcomes.

Random Vector Functional-link (RVFL) networks, as a class of random learner models, have received careful attention from the neural network research community due to their advantages in obtaining fast learning algorithms and models, in which the hidden layer parameters are randomly generated and remain fixed during the training phase. However, its universal approximation ability may not be guaranteed if the random parameters are not properly selected in an appropriate range. Moreover, the resulting random learner’s generalization performance may seriously deteriorate once the RVFL network’s structure is not well-designed. Stochastic configuration (SC) algorithm, which incrementally constructs a universal approximator by obtaining random hidden parameters under a specified supervisory mechanism, instead of fixing the selection scope in advance and without any reference to training information, can effectively circumvent these awkward issues caused by randomness. This paper extends the SC algorithm to an online sequential version, termed as an OSSC algorithm, by means of recursive least square (RLS) technique, aiming to copy with modeling tasks where training observations are sequentially provided. Compared to the online sequential learning of RVFL networks (OS-RVFL in short), our proposed OSSC algorithm can avoid the awkward setting of certain unreasonable range for the random parameters, and can also successfully build a random learner with preferable learning and generalization capabilities. The experimental study has shown the effectiveness and advantages of our OSSC algorithm.

The ability to communicate effectively in writing and produce clear and cohesive text is a necessary skill in both educational settings and the workplace, yet many young students struggle to organize their thoughts and engage in deep thinking. To address these challenges, an augmented reality (AR) application titled “Explore Wild Animals” has been used to help students organize information; however, it may not accommodate different cognitive styles. Integrating formative peer assessment (FPA) strategy into AR-based instruction can enhance knowledge construction and address diverse cognitive needs. This study, conducted from May to June 2023, empirically investigates the effects of FPA in an AR environment on the writing performance of learners with field-independent (FI) and field-dependent (FD) cognitive styles. A total of 89 fifth-grade pupils from China were randomly assigned to two groups: one group adopting FPA in an AR environment (AR-FPA), and the other group adopting FPA in a conventional PowerPoint (PPT) version 2410 environment (FPA). The results of a two-way analysis of covariance (ANCOVA) indicate that the AR-FPA group outperformed the FPA group in writing performance. Specifically, FI learners benefitted more from the AR-FPA approach, while FD learners performed better with the FPA approach. However, multiple linear regression analysis reveals that the peer feedback quality and features showed little to no significant correlation with feedback providers’ writing performance, regardless of cognitive style. These results highlight the effectiveness of integrating AR and FPA in enhancing educational outcomes, providing practical insights for promoting the sustainability of technology-enhanced learning and teaching practices.