Skip to Main content Skip to Navigation
Conference papers

A correlation-based entity embedding approach for robust entity linking

Abstract : Entity alignment is a crucial tool in knowledge discovery to reconcile knowledge from different sources. Recent state-of-the-art approaches leverage joint embedding of knowledge graphs (KGs) so that similar entities from different KGs are close in the embedded space. Whatever the joint embedding technique used, a seed set of aligned entities, often provided by (time-consuming) human expertise, is required to learn the joint KG embedding and/or a mapping between KG embeddings. In this context, a key issue is to limit the size and quality requirement for the seed. State-of-the-art methods usually learn the embedding by explicitly minimizing the distance between aligned entities from the seed and uniformly maximizing the distance for entities not in the seed. In contrast, we design a less restrictive optimization criterion that indirectly minimizes the distance between aligned entities in the seed by globally maximizing the dimension-wise correlation among all the embeddings of seed entities. Within an iterative entity alignment system, the correlation-based entity embedding function achieves state-of-the-art results and is shown to significantly increase robustness to the seed's size and accuracy. It ultimately enables fully unsupervised entity alignment using a seed automatically generated with a symbolic alignment method based on entities' names.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/hal-02999303
Contributor : Cheikh Brahim El Vaigh <>
Submitted on : Thursday, November 12, 2020 - 8:22:14 AM
Last modification on : Monday, November 23, 2020 - 11:46:54 AM

File

alignd_ICTAI2020.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02999303, version 1

Citation

Cheikh Brahim El Vaigh, François Torregrossa, Robin Allesiardo, Guillaume Gravier, Pascale Sébillot. A correlation-based entity embedding approach for robust entity linking. ICTAI 2020 - IEEE 32nd International Conference on Tools with Artificial Intelligence, Nov 2020, Virtual, United States. pp.1-6. ⟨hal-02999303⟩

Share

Metrics

Record views

56

Files downloads

87