A Novel Ensemble Method for Named Entity Recognition and Disambiguation based on Neural Network thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Novel Ensemble Method for Named Entity Recognition and Disambiguation based on Neural Network

Published on Nov 22, 20182946 Views

Named entity recognition (NER) and disambiguation (NED) are subtasks of information extraction that respectively aim to recognize named entities mentioned in text, to assign them pre-defined types, an

Related categories

Chapter list

A Novel Ensemble Method for Named Entity Recognition and Disambiguation based on Neural Network00:00
“In the spring of 1800, Napoleon and his troops crossed the Swiss Alps”00:23
Extractors01:44
Ensemble strategies03:00
Mappings to the NERD Ontology04:00
Our approach Ensemble NERD04:04
4 Kinds of feature04:26
1. Surface Form Features04:35
2. Type Features - 105:17
2. Type Features - 205:52
3. Entity Features - 106:22
3. Entity Features - 206:58
4. Score Features07:20
2 NEURAL NETWORKS07:47
1. ENNTR - 107:50
1. ENNTR - 208:03
1. ENNTR - 309:05
1. ENNTR - 409:05
2. ENND - 109:31
2. ENND - 209:43
2. ENND - 310:10
2. ENND - 410:35
Results11:04
Ground Truth Corpora11:09
Type Recognition Evaluation11:29
NED Evaluation - 112:07
NED Evaluation - 212:22
Conclusions - Main Findings12:46
Future Work13:16
Thank you!13:53