A Simplifying and Learnable Graph Convolutional Attention Network for Unsupervised Knowledge Graphs Alignment

2024-10-17

arXiv.org

AI Summary

To address the high modeling complexity and the trade-off between effectiveness and scalability in unsupervised entity alignment (EA), this paper proposes SLU, a lightweight and learnable graph convolutional attention network. Methodologically: (i) we design LCAT, a novel lightweight backbone network that significantly reduces parameter count and computational overhead; (ii) we introduce a latent-matching-driven neighborhood reconstruction mechanism to enhance structural awareness; and (iii) we formulate a consistency-driven similarity metric to improve cross-graph semantic alignment robustness. Extensive experiments across multilingual/monolingual and multi-scale datasets (15K and 100K entities) demonstrate that SLU consistently outperforms 25 state-of-the-art baselines, achieving up to a 6.4% absolute gain in Hits@1. SLU thus delivers superior efficiency, accuracy, and scalability for unsupervised EA.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

The success of current Entity Alignment (EA) task depends largely on the supervision information provided by labeled data. Considering the cost of labeled data, most supervised methods are difficult to apply in practical scenarios. Therefore, more and more works based on contrastive learning, active learning or other deep learning techniques have been developed, to solve the performance bottleneck caused by the lack of labeled data. However, the existing unsupervised EA methods still have some limitations, either their modeling complexity is high or they cannot balance the effectiveness and practicality of alignment. To overcome these issues, we propose a Simplifying and Learnable graph convolutional attention network for Unsupervised Knowledge Graphs alignment method (SLU). Specifically, we first introduce LCAT, a new and simple framework as the backbone network to model the graph structure of two KGs. Then we design a reconstruction method of relation structure based on potential matching relations for efficiently filtering invalid neighborhood information of aligned entities, to improve the usability and scalability of SLU. Impressively, a similarity function based on consistency is proposed to better measure the similarity of candidate entity pairs. Finally, we conduct extensive experiments on three datasets of different sizes (15K and 100K) and different types (cross-lingual and monolingual) to verify the superiority of SLU. Experimental results show that SLU significantly improves alignment accuracy, outperforming 25 supervised or unsupervised methods, and improving 6.4% in Hits@1 over the best baseline in the best case.

Problem

• Unsupervised alignment of knowledge graphs without labeled data
• Reducing modeling complexity while maintaining alignment effectiveness
• Improving accuracy and scalability in entity alignment tasks

Innovation

• Simplified graph convolutional attention network
• Relation structure reconstruction method
• Consistency-based similarity function