Improving Distantly-Supervised Relation Extraction Through BERT-Based Label and Instance Embeddings

Improving Distantly-Supervised Relation Extraction Through BERT-Based Label and Instance Embeddings

Distantly-supervised relation extraction (RE) is an effective method to scale RE to large corpora but suffers from noisy labels. Existing approaches try to alleviate noise through multi-instance learning and by providing additional information but manage to recognize mainly the top frequent relation...

Full description

Bibliographic Details
Main Authors: Despina Christou, Grigorios Tsoumakas
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Access
Subjects:
Relation extraction
distant supervision
BERT
label embeddings
relation attention
entity information
Online Access:https://ieeexplore.ieee.org/document/9405641/

Internet

https://ieeexplore.ieee.org/document/9405641/

Similar Items