BLINK: Better entity LINKing
Entity Linking python library that uses Wikipedia as the target knowledge base.
named-entity-recognition wikification natural-language-processing code research paper library arxiv:1911.03814

In a nutshell, BLINK uses a two stage approach for entity linking, based on fine-tuned BERT architectures. In the first stage, BLINK performs retrieval in a dense space defined by a bi-encoder that independently embeds the mention context and the entity descriptions. Each candidate is then examined more carefully with a cross-encoder, that concatenates the mention and entity text. BLINK achieves state-of-the-art results on multiple datasets.

Don't forget to tag @facebookresearch , @fabiopetroni , @martinj96 , @riedelcastro in your comment, otherwise they may not be notified.

Authors community post
Share this project
Similar projects
Transformers - Hugging Face
🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.
Fairseq-tagging
A Fairseq fork for sequence tagging/labeling tasks
BRAT: Rapid Annotation Tool
BRAT (brat rapid annotation tool) is based on the stav visualiser which was originally made in order to visualise BioNLP'11 Shared Task data.
Med7 - clinical natural language processing for EHR
Med7 is a transferable clinical natural language processing model for electronic health records, compatible with spaCy, for named-entity recognition task
Top collections