Wembedder: Wikidata entity embedding web service

TL;DR

Wembedder is a web service providing multilingual embeddings for over 600,000 Wikidata entities, trained using graph walks and accessible via a REST API, facilitating easier access to structured knowledge representations.

Contribution

It introduces a publicly accessible web service for Wikidata entity embeddings, integrating graph-based training with multilingual support and API access.

Findings

Supports over 600,000 Wikidata items and properties

Uses Gensim Word2Vec with graph walks for embedding training

Provides a REST API for easy access

Abstract

I present a web service for querying an embedding of entities in the Wikidata knowledge graph. The embedding is trained on the Wikidata dump using Gensim's Word2Vec implementation and a simple graph walk. A REST API is implemented. Together with the Wikidata API the web service exposes a multilingual resource for over 600'000 Wikidata items and properties.