Commonsense Knowledge Graph Completion Via Contrastive Pretraining and Node Clustering
This work addresses challenges in CSKG completion, which is important for AI systems requiring commonsense reasoning, but it appears incremental as it builds on existing methods with specific enhancements.
The paper tackles the problems of edge sparsity and node redundancy in commonsense knowledge graphs (CSKGs) by proposing a framework called CPNC, which uses contrastive pretraining and node clustering to improve node representation and concept aggregation, resulting in state-of-the-art performance on benchmarks like CN-100K and ATOMIC.
The nodes in the commonsense knowledge graph (CSKG) are normally represented by free-form short text (e.g., word or phrase). Different nodes may represent the same concept. This leads to the problems of edge sparsity and node redundancy, which challenges CSKG representation and completion. On the one hand, edge sparsity limits the performance of graph representation learning; On the other hand, node redundancy makes different nodes corresponding to the same concept have inconsistent relations with other nodes. To address the two problems, we propose a new CSKG completion framework based on Contrastive Pretraining and Node Clustering (CPNC). Contrastive Pretraining constructs positive and negative head-tail node pairs on CSKG and utilizes contrastive learning to obtain better semantic node representation. Node Clustering aggregates nodes with the same concept into a latent concept, assisting the task of CSKG completion. We evaluate our CPNC approach on two CSKG completion benchmarks (CN-100K and ATOMIC), where CPNC outperforms the state-of-the-art methods. Extensive experiments demonstrate that both Contrastive Pretraining and Node Clustering can significantly improve the performance of CSKG completion. The source code of CPNC is publicly available on \url{https://github.com/NUSTM/CPNC}.