Deep Active Alignment of Knowledge Graph Entities and Schemata

TL;DR

This paper introduces DAAKG, a deep learning and active learning-based method for aligning entities, relations, and classes across knowledge graphs, improving accuracy and efficiency in schema and entity alignment.

Contribution

The paper presents a novel semi-supervised deep learning approach combined with active learning for comprehensive knowledge graph alignment at multiple levels.

Findings

DAAKG achieves superior accuracy on benchmark datasets.

The method effectively aligns entities, relations, and classes.

All modules of DAAKG are validated through experiments.

Abstract

Knowledge graphs (KGs) store rich facts about the real world. In this paper, we study KG alignment, which aims to find alignment between not only entities but also relations and classes in different KGs. Alignment at the entity level can cross-fertilize alignment at the schema level. We propose a new KG alignment approach, called DAAKG, based on deep learning and active learning. With deep learning, it learns the embeddings of entities, relations and classes, and jointly aligns them in a semi-supervised manner. With active learning, it estimates how likely an entity, relation or class pair can be inferred, and selects the best batch for human labeling. We design two approximation algorithms for efficient solution to batch selection. Our experiments on benchmark datasets show the superior accuracy and generalization of DAAKG and validate the effectiveness of all its modules.