Combining Word and Entity Embeddings for Entity Linking thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Combining Word and Entity Embeddings for Entity Linking

Published on Jul 10, 2017854 Views

The correct identification of the link between an entity mention in a text and a known entity in a large knowledge base is important in information retrieval or information extraction. The general app

Related categories

Chapter list

Combining word and entity embeddings for entity linking00:00
Entity linking00:21
Entity disambiguation01:26
Outline02:12
Preliminaries02:34
What can we use?03:44
From word to entity embeddings04:45
Our Extended Anchor Text model07:30
Using EAT08:47
Main advantages10:11
Common Entity Linking architecture10:44
Baseline features11:15
EAT features12:55
Candidate representation14:16
Candidate Ranking (Training)15:13
Candidate Ranking (Prediction)16:22
Evaluation setup17:34
Entity linking results18:58
Entity linking results - 120:23
Conclusions21:46
Untitled23:27