Coreference resolution systems can benefit greatly from inclusion of global context, and a number of recent approaches have demonstrated improvements when precomputing an alignment to external knowledge sources. However, since alignment itself is a challenging task and is often noisy, existing systems either align conservatively, resulting in very few links, or combine the attributes of multiple candidates, leading to a conflation of entities. Our approach instead maintains ranked lists of candidate entities that are dynamically merged and reranked during inference. Further, we incorporate a large set of surface string variations for each entity by using anchor texts from the web that link to the entity. These forms of global context enable our system to outperform a competitive baseline without a knowledge base by 1.09 B3 F1 points, and a state-of-the-art system by 0.41 points on the ACE 2004 data.
Dynamic Knowledge-Base Alignment for Coreference Resolution
1. Dynamic Knowledge-Base Alignment for
Coreference Resolution
Jiaping Zheng, LukeVilnis, Sameer Singh,
Jinho D. Choi,Andrew McCallum
Presented at CoNLL 2013
University of Massachusetts Amherst
Thursday, August 8, 13
2. Coreference Resolution
2
The Chicago suburb of Arlington Heights is the first stop
for George W. Bush today. The Texas governor stops in
Gore’s home state of Tennessee this afternoon ...
Thursday, August 8, 13
3. Coreference Resolution
• Identify mentions that refer to the same entity.
2
The Chicago suburb of Arlington Heights is the first stop
for George W. Bush today. The Texas governor stops in
Gore’s home state of Tennessee this afternoon ...
Thursday, August 8, 13
4. Coreference Resolution
• Identify mentions that refer to the same entity.
2
The Chicago suburb of Arlington Heights is the first stop
for George W. Bush today. The Texas governor stops in
Gore’s home state of Tennessee this afternoon ...
Thursday, August 8, 13
5. Coreference Resolution
• Identify mentions that refer to the same entity.
2
The Chicago suburb of Arlington Heights is the first stop
for George W. Bush today. The Texas governor stops in
Gore’s home state of Tennessee this afternoon ...
Thursday, August 8, 13
6. Coreference Resolution
• Identify mentions that refer to the same entity.
• Useful in relation extraction, question answering,
machine translation, etc.
2
The Chicago suburb of Arlington Heights is the first stop
for George W. Bush today. The Texas governor stops in
Gore’s home state of Tennessee this afternoon ...
Thursday, August 8, 13
8. Coreference Resolution
• Determine the entity in a reference knowledge-base for
textual mentions.
3
… George W. !
Bush ...!
Thursday, August 8, 13
9. Coreference Resolution
• Determine the entity in a reference knowledge-base for
textual mentions.
3
… George W. !
Bush ...!
Entity! Attr!
George W. Bush! ...!
...!
Thursday, August 8, 13
12. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
4
Thursday, August 8, 13
13. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
- High precision, but fewer alignments.
4
Thursday, August 8, 13
14. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
- High precision, but fewer alignments.
- Ponzetto & Strube 2006.
- Ratinov & Roth 2012.
4
Thursday, August 8, 13
15. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
- High precision, but fewer alignments.
- Ponzetto & Strube 2006.
- Ratinov & Roth 2012.
• Linking a mention to multiple knowledge-base entities.
4
Thursday, August 8, 13
16. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
- High precision, but fewer alignments.
- Ponzetto & Strube 2006.
- Ratinov & Roth 2012.
• Linking a mention to multiple knowledge-base entities.
- Higher recall, but conflates entities.
4
Thursday, August 8, 13
17. Entity Linking
• Provides global context for coreference resolution.
• Linking a mention to one knowledge-base entity.
- High precision, but fewer alignments.
- Ponzetto & Strube 2006.
- Ratinov & Roth 2012.
• Linking a mention to multiple knowledge-base entities.
- Higher recall, but conflates entities.
- Rahman & Ng 2011.
4
Thursday, August 8, 13
19. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
5
Thursday, August 8, 13
20. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
5
Thursday, August 8, 13
21. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
- Querying knowledge-based bridge (Dalton & Dietz, 2013).
5
Thursday, August 8, 13
22. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
- Querying knowledge-based bridge (Dalton & Dietz, 2013).
2. Merge entity lists when mentions are coreferenced.
5
Thursday, August 8, 13
23. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
- Querying knowledge-based bridge (Dalton & Dietz, 2013).
2. Merge entity lists when mentions are coreferenced.
3. Re-rank the merged list.
5
Thursday, August 8, 13
24. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
- Querying knowledge-based bridge (Dalton & Dietz, 2013).
2. Merge entity lists when mentions are coreferenced.
3. Re-rank the merged list.
4. Attributes are extracted from the top ranked entity.
5
Thursday, August 8, 13
25. Dynamic Alignment
1. Compute initial ranked list of knowledge-base entities
for named entities.
- List of Wikipedia articles.
- Querying knowledge-based bridge (Dalton & Dietz, 2013).
2. Merge entity lists when mentions are coreferenced.
3. Re-rank the merged list.
4. Attributes are extracted from the top ranked entity.
- Surface string variations from the web.
5
Thursday, August 8, 13
26. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Thursday, August 8, 13
27. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Thursday, August 8, 13
28. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Thursday, August 8, 13
29. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Washington State
Thursday, August 8, 13
30. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Washington State
Thursday, August 8, 13
31. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Washington State
Washington, DC
Washington State
Car Wash
The Wash
Thursday, August 8, 13
32. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Washington State
Washington, DC
Washington State
Car Wash
The Wash
Thursday, August 8, 13
33. Dynamic Alignment
6
… about navigation charts
that he had ordered from a
company based in
Washington …
… opened one of them to
discover the absentee ballot
of Steven H. Forrester of
Bellevue, Wash. …
… were not meaningful
because counting in
Washington State has
been completed …
Washington, DC
Washington State
Car Wash
The Wash
Washington State
Washington, DC
Washington State
Car Wash
The WashWashington State
Washington, DC
Car Wash
The Wash
Thursday, August 8, 13
36. Experiments
• ACE 2004 dataset.
• Baseline.
- Pairwise classification system.
- No external knowledge sources.
- A rich set of features.
- Best link strategy.
- L2-regularized SVM using hinge-loss.
7
Thursday, August 8, 13
37. Experiments
• ACE 2004 dataset.
• Baseline.
- Pairwise classification system.
- No external knowledge sources.
- A rich set of features.
- Best link strategy.
- L2-regularized SVM using hinge-loss.
• Static linking.
7
Thursday, August 8, 13
38. Experiments
• ACE 2004 dataset.
• Baseline.
- Pairwise classification system.
- No external knowledge sources.
- A rich set of features.
- Best link strategy.
- L2-regularized SVM using hinge-loss.
• Static linking.
• Dynamic linking.
7
Thursday, August 8, 13
41. Experiments
10
un
di-
on-
ns
ti-
in-
lu-
ise
ith
our
5510 15 20 25 30 35 40 45 50
0.6
0.8
1
1.2
1.4
1.6
1.8
Top X% of Docs by Number of Mentions
ImprovementoverBaseline
Dynamic Linking
Static Linking
Figure 2: Improvements on the top X% of docu-
Thursday, August 8, 13
42. Conclusion
• Coreference resolution systems benefit greatly from
inclusion of global context.
• Linking mentions to a knowledge base provides this
context.
• Maintaining a ranked list of entities outperforms
previous fixed alignment approaches.
11
Thursday, August 8, 13