KBP2014 Entity Linking Scorer B-cubed++

14
KBP2014 Entity Linking Scorer B- cubed++ Xiaoman Pan, Qi Li and Heng Ji {panx2,liq7,jih}rpi.edu

description

KBP2014 Entity Linking Scorer B-cubed++. Xiaoman Pan, Qi Li and Heng Ji {panx2,liq7,jih}rpi.edu. Overview. We augment the B-Cubed++ metric to evaluate the KBP2014 Entity Discovery and Linking results . - PowerPoint PPT Presentation

Transcript of KBP2014 Entity Linking Scorer B-cubed++

Page 1: KBP2014 Entity Linking Scorer B-cubed++

KBP2014 Entity Linking Scorer B-cubed++

Xiaoman Pan, Qi Li and Heng Ji

{panx2,liq7,jih}rpi.edu

Page 2: KBP2014 Entity Linking Scorer B-cubed++

Overview

We augment the B-Cubed++ metric to evaluate the KBP2014 Entity Discovery and Linking results.

We use a tuple doc-id, start, end, entity-type, kb-id⟨ ⟩

to represent each entity mention, where a special type of kb-id is NIL.

Page 3: KBP2014 Entity Linking Scorer B-cubed++

Overview

Let s be an entity mention in the system output, g be an corresponding gold-standard. An output mention s matches a reference mention g iff:

1. s.doc-id = g.doc-id,2. s.start = g.start, s.end = g.end, 3. s.entity-type = g.entity-type,4. and s.kb-id = g.kb-id.

Page 4: KBP2014 Entity Linking Scorer B-cubed++

Overview

Page 5: KBP2014 Entity Linking Scorer B-cubed++

Overview

Page 6: KBP2014 Entity Linking Scorer B-cubed++
Page 7: KBP2014 Entity Linking Scorer B-cubed++
Page 8: KBP2014 Entity Linking Scorer B-cubed++
Page 9: KBP2014 Entity Linking Scorer B-cubed++
Page 10: KBP2014 Entity Linking Scorer B-cubed++
Page 11: KBP2014 Entity Linking Scorer B-cubed++

Another NIL example

Page 12: KBP2014 Entity Linking Scorer B-cubed++
Page 13: KBP2014 Entity Linking Scorer B-cubed++
Page 14: KBP2014 Entity Linking Scorer B-cubed++