What is it about?
A state-of-the-art, modular entity linking package that is efficient and easy to use as a python package or Web API. REL utilizes English Wikipedia as a knowledge base and can be used for Entity linking (given a text, the system outputs a list of mention-entity pairs, where each mention is a n-gram from text and each entity is an entity in the knowledge base) and Entity Disambiguation (given a text and a list of mentions, the system assigns an entity or NIL to each mention).
Featured Image
Photo by James on Unsplash
Why is it important?
Entity Linking is a important basic step to enrich documents, but for years, only outdated systems have been available for practical application beyond a pure research setting. REL integrates the key improvements from the last decade of research in this domain and makes these accessible to a wide audience.
Perspectives
Entity Linking is a quite complex task, but potentially useful for many applications in the domains of search and document analysis. REL provides researchers in these domains with state-of-the-art entity annotations, without much effort.
Prof.dr.ir. Arjen P. de Vries
Radboud Universiteit
Read the Original
This page is a summary of: REL: An Entity Linker Standing on the Shoulders of Giants, July 2020, ACM (Association for Computing Machinery),
DOI: 10.1145/3397271.3401416.
You can read the full text:
Resources
Contributors
The following have contributed to this page
