English

$\mu\text{KG}$: A Library for Multi-source Knowledge Graph Embeddings and Applications

Computation and Language 2022-07-29 v2 Artificial Intelligence Machine Learning

Abstract

This paper presents μKG\mu\text{KG}, an open-source Python library for representation learning over knowledge graphs. μKG\mu\text{KG} supports joint representation learning over multi-source knowledge graphs (and also a single knowledge graph), multiple deep learning libraries (PyTorch and TensorFlow2), multiple embedding tasks (link prediction, entity alignment, entity typing, and multi-source link prediction), and multiple parallel computing modes (multi-process and multi-GPU computing). It currently implements 26 popular knowledge graph embedding models and supports 16 benchmark datasets. μKG\mu\text{KG} provides advanced implementations of embedding techniques with simplified pipelines of different tasks. It also comes with high-quality documentation for ease of use. μKG\mu\text{KG} is more comprehensive than existing knowledge graph embedding libraries. It is useful for a thorough comparison and analysis of various embedding models and tasks. We show that the jointly learned embeddings can greatly help knowledge-powered downstream tasks, such as multi-hop knowledge graph question answering. We will stay abreast of the latest developments in the related fields and incorporate them into μKG\mu\text{KG}.

Keywords

Cite

@article{arxiv.2207.11442,
  title  = {$\mu\text{KG}$: A Library for Multi-source Knowledge Graph Embeddings and Applications},
  author = {Xindi Luo and Zequn Sun and Wei Hu},
  journal= {arXiv preprint arXiv:2207.11442},
  year   = {2022}
}

Comments

Accepted in the 21th International Semantic Web Conference (ISWC 2022)

R2 v1 2026-06-25T01:09:57.724Z