Nested Named Entity Recognition via Second-best Sequence Learning and Decoding

5 Sep 2019 Takashi Shibuya Eduard Hovy

When an entity name contains other names within it, the identification of all combinations of names can become difficult and expensive. We propose a new method to recognize not only outermost named entities but also inner nested ones... (read more)

PDF Abstract

Datasets


Results from the Paper


 Ranked #1 on Nested Named Entity Recognition on ACE 2004 (using extra training data)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
RESULT BENCHMARK
Nested Named Entity Recognition ACE 2004 Second-best learning and decoding + BERT + Flair F1 85.82 # 1
Nested Named Entity Recognition ACE 2004 Second-best learning and decoding + BERT F1 84.97 # 2
Named Entity Recognition ACE 2004 Second-best learning and decoding F1 85.82 # 2
Nested Mention Recognition ACE 2004 Second-best learning and decoding F1 85.82 # 2
Nested Named Entity Recognition ACE 2004 Second-best learning and decoding F1 77.44 # 5
Nested Named Entity Recognition ACE 2005 Second-best learning and decoding + BERT + Flair F1 84.34 # 1
Nested Mention Recognition ACE 2005 Second-best learning and decoding F1 84.34 # 1
Nested Named Entity Recognition ACE 2005 Second-best learning and decoding F1 76.83 # 6
Nested Named Entity Recognition ACE 2005 Second-best learning and decoding + BERT F1 83.99 # 3
Named Entity Recognition ACE 2005 Second-best learning and decoding F1 84.34 # 3
Nested Named Entity Recognition GENIA Second-best learning and decoding + BERT F1 77.05 # 7
Named Entity Recognition GENIA Second-best learning and decoding F1 77.19 # 4
Nested Named Entity Recognition GENIA Second-best learning and decoding + BERT + Flair F1 77.36 # 5
Nested Named Entity Recognition GENIA Second-best learning and decoding F1 77.19 # 6
Named Entity Recognition GENIA Second-best learning and decoding + BERT + Flair F1 77.36 # 3

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet