Anaphora Resolution

1
Anaphora Resolution

• Sanghoon Kwak
• Takahiro Aoyama

2
Anaphora Resolution

• Moussa said the presidents discussed the
  situation in Iraq and would return to it Monday.
• However, he would not comment on whether Syria
  and Egypt were planning to send planes to
  Baghdad.
• Anaphora resolution is the process of determining
  whether a noun phrase refers to an antecedent
  noun phrase.

3
Approach

• Extracted 12 features for each noun phrase pair
  from a news article corpora
• Constructed a training set using these 12
  features. Used 70 news articles from NY Times
  and Broadcast News.
• Built a decision tree with this training data
  using the C4.5 decision tree builder

4
Features

• Some of 12 features include
• Distance
• Distance by the number of sentences
• Pronoun
• Whether a noun phrase is a pronoun or not
• String match
• Whether two phrases have the same character
  sequence
• Semantic Class
• Whether two phrases are in the same semantic
  class (Person, Organization, etc.)
• Number
• Whether two phrases are both plural or both
  singular

5
Result

• Training and Test Result

Training Testing
Phrase Pairs 406,400 212,865
Incorrectly Classified 24,762 14,213
Error Rate 6.1 6.7

• Rules based on decision tree
• DIST lt 0, JPRO True, NUM True, SEMCLASS
  True
• ? Coreferent
• IPRO True, STR MATCH true, DIST gt 0
• ? Not Coreferent

6
Future Work

• Train and test data from different domains
• (Blog posts, E-Mail messages, etc.)
• Further subdivide decision tree classes