2020年6月13日 · To solve the above problems, we propose a hierarchical graph knowledge tracing model called HGKT to explore the latent hierarchical relations between exercises. Specifically, we introduce the concept of problem schema to construct a hierarchical exercise graph that could model the exercise learning dependencies.

The implementation of the paper Introducing Problem Schema with Hierarchical Exercise Graph for Knowledge Tracing. The architecture of the HGKT is as follows: To run this code you need the following:

为了解决上述问题，我们提出了一个叫做HGKT的分层图知识追踪模型，以探索练习之间的潜在分层关系。 具体来说，我们引入了问题模式的概念来构建一个层次化的练习图，该图可以模拟练习的学习依赖关系。 此外，我们采用两种关注机制来突出学习者的重要历史状态。 在测试阶段，我们提出了一个K&S诊断矩阵，可以追踪知识和问题模式的掌握情况，这可以更容易地应用于不同的应用程序。 广泛的实验显示了我们提出的模型的有效性和可解释性。 知识追踪是计算机辅助教 …

In this paper, a novel hypergraph-based knowledge tracing model (HGKT) is proposed to address these limitations. Firstly, we exploit edge feature that indicates the frequency of exercise-concept's occurrence to extend the common bipartite graph.

To solve the above problems, we propose a hierarchical graph knowledge trac-ing model called HGKT to explore the latent hierarchical relations between exercises. Specifically, we introduce the concept of prob-lem schema to construct a hierarchical exercise graph that could model the exercise learning dependencies.

基于图的知识跟踪 (GKT)将知识跟踪与图神经网络 [25]相结合。 它将学习者隐藏的知识状态编码为图节点的嵌入，并更新知识图中的状态。 这些模型已被证明是有效的，但仍有局限性。 现有的方法由于没有考虑练习的文本，都面临着练习表征缺失的问题。 对于基于练习的轨迹，据我们所知，练习增强知识追踪 (exercise Enhanced knowledge Tracing, EKT) 是第一个将练习文本的特征整合到知识追踪模型 [16]中的方法。 而EKT通过将练习文本直接输入双向LSTM网络 [14]来提取 …

2022年7月7日 · This video shows how we propose a hierarchical graph knowledge tracing model called HGKT to explore the latent complex relations between exercises. Specifically, we introduce the concept of problem schema to construct a hierarchical exercise graph that could model the exercise learning dependencies.

2020年6月13日 · To solve the above problems, we propose a hierarchical graph knowledge tracing model called HGKT to explore the latent hierarchical relations between exercises. Specifically, we introduce the concept of problem schema to construct a hierarchical exercise graph that could model the exercise learning dependencies.

knowledge tracing framework named HGKT, which uni•es the strengths of hierarchical graph neural network and recurrent se-quence model with a−ention to enhance the performance of knowl-edge tracing. In HKGT, we present an approximate method to get the problem schema features: many successes have demonstrated

A novel hypergraph-based knowledge tracing model (HGKT) is proposed that outperforms previous classical methods in terms of AUC on the three widely used datasets and the difficulty of exercises and the average response time are utilized to …