Abstract:Entity alignment (EA) is to identify equivalent entities across different knowledge graphs (KGs), which can help fuse these KGs into a more comprehensive one. Previous EA methods mainly focus on aligning a pair of KGs, and to the best of our knowledge, no existing EA method considers aligning multiple (more than two) KGs. To fill this research gap, in this work, we study a novel problem of aligning multiple KGs and propose an effective framework named MultiEA to solve the problem. First, we embed the entities of all the candidate KGs into a common feature space by a shared KG encoder. Then, we explore three alignment strategies to minimize the distances among pre-aligned entities. In particular, we propose an innovative inference enhancement technique to improve the alignment performance by incorporating high-order similarities. Finally, to verify the effectiveness of MultiEA, we construct two new real-world benchmark datasets and conduct extensive experiments on them. The results show that our MultiEA can effectively and efficiently align multiple KGs in a single pass.

What problem does this paper attempt to address?

### Problems Addressed by the Paper This paper aims to address a new problem in the field of Entity Alignment (EA), specifically aligning multiple (more than two) Knowledge Graphs (KGs) in a single operation. Existing research primarily focuses on the alignment between a pair of KGs, while this paper attempts to solve the problem of aligning multiple KGs simultaneously for the first time. #### Main Contributions 1. **Proposing a New Problem**: The authors are the first to propose the problem of aligning multiple (more than two) KGs simultaneously and have constructed corresponding benchmark datasets to support this research. 2. **Introducing the MultiEA Framework**: An effective framework called MultiEA is designed to address the multi-KG alignment problem by exploring three different alignment strategies and an innovative inference enhancement technique to improve alignment performance. 3. **Experimental Validation**: Extensive experiments on two newly constructed benchmark datasets demonstrate that MultiEA can effectively align multiple KGs in a single step. The authors plan to open-source the code and datasets to facilitate further research. #### Research Background - Knowledge Graphs (KGs) store a large amount of structured facts about the real world, with each fact typically represented as a triple (head entity, relation, tail entity). - Existing EA methods mainly target the alignment between two KGs. These methods cannot be directly applied to the problem of aligning multiple KGs because they can only utilize local (pairwise) alignment information and fail to capture global (beyond pairwise) alignment information. - This paper captures useful global alignment information by embedding entities from all candidate KGs into a unified space, which helps achieve better alignment performance. #### Method Overview - **KG Encoding**: A shared GNN encoder is used to embed entities from all candidate KGs into a common vector space. - **Alignment Strategies**: Three alignment strategies are proposed to minimize the distance between pre-aligned entities. - **Inference Enhancement**: An innovative inference enhancement technique is developed to significantly improve alignment performance by incorporating higher-order similarity. #### Experimental Results - Experimental results on two benchmark datasets show that MultiEA can efficiently align multiple KGs in a single operation, validating its effectiveness and efficiency. In summary, this paper fills a research gap in the field of multi-KG alignment and proposes a new framework, MultiEA, to address this challenging problem.

Aligning Multiple Knowledge Graphs in a Single Pass

Understanding and Improving Knowledge Graph Embedding for Entity Alignment

LargeEA: Aligning Entities for Large-scale Knowledge Graphs

Large-scale Entity Alignment Via Knowledge Graph Merging, Partitioning and Embedding

Multi-view Knowledge Graph Embedding for Entity Alignment

AEP: Aligning Knowledge Graphs Via Embedding Propagation

A survey: knowledge graph entity alignment research based on graph embedding

Collective Entity Alignment for Knowledge Fusion of Power Grid Dispatching Knowledge Graphs

DERA: Dense Entity Retrieval for Entity Alignment in Knowledge Graphs

Cross-lingual knowledge graph entity alignment based on relation awareness and attribute involvement

Two Heads Are Better Than One: Integrating Knowledge from Knowledge Graphs and Large Language Models for Entity Alignment

Iterative Entity Alignment via Joint Knowledge Embeddings

A Comprehensive Survey of Entity Alignment for Knowledge Graphs

Vision, Deduction and Alignment: An Empirical Study on Multi-modal Knowledge Graph Alignment

LLM-Align: Utilizing Large Language Models for Entity Alignment in Knowledge Graphs

MMEA: Entity Alignment for Multi-modal Knowledge Graph.

Entity Alignment with Global Information Aggregation

Toward Practical Entity Alignment Method Design: Insights from New Highly Heterogeneous Knowledge Graph Datasets

An effective knowledge graph entity alignment model based on multiple information

Knowledge Graph Entity Alignment Using Relation Structural Similarity

A benchmarking study of embedding-based entity alignment for knowledge graphs