DECRL: A Deep Evolutionary Clustering Jointed Temporal Knowledge Graph Representation Learning Approach
Qian Chen, Ling Chen
TL;DR
DECRL addresses the challenge of capturing the temporal evolution of high-order correlations in temporal knowledge graphs by integrating deep evolutionary clustering with TKG representation learning. It introduces a cluster-aware one-to-one alignment via Hungarian matching, an implicit cluster-correlation encoder guided by a global graph, a time residual gate, and an attentive temporal encoder, all feeding into a ConvTransE decoder for event prediction. Empirical results on seven real-world datasets demonstrate state-of-the-art performance with significant gains over strong baselines in MRR and Hits@k, validating the effectiveness of modeling evolving high-order dependencies. The approach enhances the practical utility of TKGs for predictive tasks in domains such as information retrieval, governance, and safety analyses by providing more accurate and temporally coherent embeddings of entities and their relations.
Abstract
Temporal Knowledge Graph (TKG) representation learning aims to map temporal evolving entities and relations to embedded representations in a continuous low-dimensional vector space. However, existing approaches cannot capture the temporal evolution of high-order correlations in TKGs. To this end, we propose a Deep Evolutionary Clustering jointed temporal knowledge graph Representation Learning approach (DECRL). Specifically, a deep evolutionary clustering module is proposed to capture the temporal evolution of high-order correlations among entities. Furthermore, a cluster-aware unsupervised alignment mechanism is introduced to ensure the precise one-to-one alignment of soft overlapping clusters across timestamps, thereby maintaining the temporal smoothness of clusters. In addition, an implicit correlation encoder is introduced to capture latent correlations between any pair of clusters under the guidance of a global graph. Extensive experiments on seven real-world datasets demonstrate that DECRL achieves the state-of-the-art performances, outperforming the best baseline by an average of 9.53%, 12.98%, 10.42%, and 14.68% in MRR, Hits@1, Hits@3, and Hits@10, respectively.