TY - JOUR
T1 - Multiscale Weisfeiler-Leman Directed Graph Neural Networks for Prerequisite-Link Prediction
AU - Zhang, Yupei
AU - Qu, Xiran
AU - Liu, Shuhui
AU - Pang, Yan
AU - Shang, Xuequn
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Prerequisite-link Prediction (PLP) aims to discover the condition relations of a specific event or a concerned variable, which is a fundamental problem in a large number of fields, such as educational data mining. Current studies on PLP usually developed graph neural networks (GNNs) to learn the representations of pairs of nodes. However, these models fail to distinguish non-isomorphic graphs and integrate multiscale structures, leading to the insufficient expressive capability of GNNs. To this end, we in this paper proposed k-dimensional Weisferiler-Leman directed GNNs, dubbed k-WediGNNs, to recognize non-isomorphic graphs via the Weisferiler-Leman algorithm. Furthermore, we integrated the multiscale structures of a directed graph into k-WediGNNs, dubbed multiscale k-WediGNNs, from the bidirected views of in-degree and out-degree. With the Siamese network, the proposed models are extended to address the problem of PLP. Besides, the expressive power is then interpreted via theoretical proofs. The experiments were conducted on four publicly available datasets for concept prerequisite relation prediction (CPRP). The results show that the proposed models achieve better performance than the state-of-the-art approaches, where our multiscale k-WediGNN achieves a new benchmark in the task of CPRP.
AB - Prerequisite-link Prediction (PLP) aims to discover the condition relations of a specific event or a concerned variable, which is a fundamental problem in a large number of fields, such as educational data mining. Current studies on PLP usually developed graph neural networks (GNNs) to learn the representations of pairs of nodes. However, these models fail to distinguish non-isomorphic graphs and integrate multiscale structures, leading to the insufficient expressive capability of GNNs. To this end, we in this paper proposed k-dimensional Weisferiler-Leman directed GNNs, dubbed k-WediGNNs, to recognize non-isomorphic graphs via the Weisferiler-Leman algorithm. Furthermore, we integrated the multiscale structures of a directed graph into k-WediGNNs, dubbed multiscale k-WediGNNs, from the bidirected views of in-degree and out-degree. With the Siamese network, the proposed models are extended to address the problem of PLP. Besides, the expressive power is then interpreted via theoretical proofs. The experiments were conducted on four publicly available datasets for concept prerequisite relation prediction (CPRP). The results show that the proposed models achieve better performance than the state-of-the-art approaches, where our multiscale k-WediGNN achieves a new benchmark in the task of CPRP.
KW - Prerequisite link prediction
KW - Weisfeiler-Leman test
KW - directed graph neural networks
KW - multiscale structure
UR - http://www.scopus.com/inward/record.url?scp=105000664876&partnerID=8YFLogxK
U2 - 10.1109/TKDE.2025.3552045
DO - 10.1109/TKDE.2025.3552045
M3 - 文章
AN - SCOPUS:105000664876
SN - 1041-4347
VL - 37
SP - 3556
EP - 3569
JO - IEEE Transactions on Knowledge and Data Engineering
JF - IEEE Transactions on Knowledge and Data Engineering
IS - 6
ER -