Matching Entities Across Different Knowledge Graphs with Graph Embeddings
This work addresses entity matching for knowledge graph integration, providing datasets and baselines, but it is incremental as it applies existing methods to new data.
The paper tackles the problem of matching entities across different knowledge graphs, such as DBpedia and Wikidata, by using a classification-based approach with RDF2Vec graph embeddings, achieving high accuracy with minimal training data on datasets of several hundred thousand ambiguous entities.
This paper explores the problem of matching entities across different knowledge graphs. Given a query entity in one knowledge graph, we wish to find the corresponding real-world entity in another knowledge graph. We formalize this problem and present two large-scale datasets for this task based on exiting cross-ontology links between DBpedia and Wikidata, focused on several hundred thousand ambiguous entities. Using a classification-based approach, we find that a simple multi-layered perceptron based on representations derived from RDF2Vec graph embeddings of entities in each knowledge graph is sufficient to achieve high accuracy, with only small amounts of training data. The contributions of our work are datasets for examining this problem and strong baselines on which future work can be based.