Corpus ID: 235458246

A Self-supervised Method for Entity Alignment

  title={A Self-supervised Method for Entity Alignment},
  author={Xiao Liu and Haoyun Hong and Xinghao Wang and Zeyi Chen and E. Kharlamov and Yuxiao Dong and Jie Tang},
Entity alignment, aiming to identify equivalent entities across different knowledge graphs (KGs), is a fundamental problem for constructing large-scale KGs. Over the course of its development, supervision has been considered necessary for accurate alignments. Inspired by the recent progress of self-supervised learning, we explore the extent to which we can get rid of supervision for entity alignment. Existing supervised methods for this task focus on pulling each pair of positive (labeled… Expand

Figures and Tables from this paper


Jointly Learning Entity and Relation Representations for Entity Alignment
This paper presents a novel joint learning framework for entity alignment that is a Graph Convolutional Network (GCN) based framework for learning both entity and relation representations and incorporates the relation approximation into entities to iteratively learn better representations for both. Expand
Relation-Aware Entity Alignment for Heterogeneous Knowledge Graphs
A novel Relation-aware Dual-Graph Convolutional Network is proposed to incorporate relation information via attentive interactions between the knowledge graph and its dual relation counterpart, and further capture neighboring structures to learn better entity representations. Expand
COTSAE: CO-Training of Structure and Attribute Embeddings for Entity Alignment
This work proposes COTSAE that combines the structure and attribute information of entities by co-training two embedding learning components, respectively, and proposes a joint attention method in the model to learn the attentions of attribute types and values cooperatively. Expand
Modeling Multi-mapping Relations for Precise Cross-lingual Entity Alignment
Experimental results show that this approach significantly outperforms many other embedding-based approaches with state-of-the-art performance and proposes a weighted negative sampling strategy to generate valuable negative samples during training and regard prediction as a bidirectional problem in the end. Expand
Iterative Entity Alignment via Joint Knowledge Embeddings
This paper presents a novel approach for entity alignment via joint knowledge embeddings that jointly encodes both entities and relations of various KGs into a unified low-dimensional semantic space according to a small seed set of aligned entities. Expand
Semi-supervised Entity Alignment via Joint Knowledge Embedding Model and Cross-graph Model
This paper proposes a semi-supervised entity alignment method by joint Knowledge Embedding model and Cross-Graph model (KECG), which can make better use of seed alignments to propagate over the entire graphs with KG-based constraints. Expand
Multi-view Knowledge Graph Embedding for Entity Alignment
A novel framework is proposed that unifies multiple views of entities to learn embeddings for entity alignment, and significantly outperforms the state-of-the-art embedding-based entity alignment methods. Expand
Semi-Supervised Entity Alignment via Knowledge Graph Embedding with Awareness of Degree Difference
This paper proposes a semi-supervised entity alignment method (SEA) to leverage both labeled entities and the abundant unlabeled entity information for the alignment, and improves the knowledge graph embedding with awareness of the degree difference by performing the adversarial training. Expand
Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs
Recurrent skipping networks (RSNs) are proposed, which employ a skipping mechanism to bridge the gaps between entities and outperformed state-of-the-art embedding-based methods for entity alignment and achieved competitive performance for KG completion. Expand
Multi-Channel Graph Neural Network for Entity Alignment
A novel Multi-channel Graph Neural Network model (MuGNN) is proposed to learn alignment-oriented knowledge graph (KG) embeddings by robustly encoding two KGs via multiple channels and infer and transfer rule knowledge for completing two KG consistently. Expand