Graph Collaborative Attention Network for Link Prediction in Knowledge Graphs
This work addresses knowledge graph completion tasks, which are important for applications like information retrieval and automated reasoning, but it is incremental as it refines an existing attention-based architecture.
The paper tackled link prediction in knowledge graphs by introducing GCAT, a refined graph neural network model that enhances context aggregation and interaction between heterogeneous nodes, achieving competitive or superior performance compared to existing neural embedding models on four benchmark datasets.
Knowledge graphs offer a structured representation of real-world entities and their relationships, enabling a wide range of applications from information retrieval to automated reasoning. In this paper, we conduct a systematic comparison between traditional rule-based approaches and modern deep learning methods for link prediction. We focus on KBGAT, a graph neural network model that leverages multi-head attention to jointly encode both entity and relation features within local neighborhood structures. To advance this line of research, we introduce \textbf{GCAT} (Graph Collaborative Attention Network), a refined model that enhances context aggregation and interaction between heterogeneous nodes. Experimental results on four widely-used benchmark datasets demonstrate that GCAT not only consistently outperforms rule-based methods but also achieves competitive or superior performance compared to existing neural embedding models. Our findings highlight the advantages of attention-based architectures in capturing complex relational patterns for knowledge graph completion tasks.