private void ExtractPronounForHeadword(IndexedWord headword, SemanticGraph dep, ICoreMap s, IList <Mention> mentions, ICollection <IntPair> mentionSpanSet, ICollection <IntPair> namedEntitySpanSet)
{
IList <CoreLabel> sent = s.Get(typeof(CoreAnnotations.TokensAnnotation));
SemanticGraph basic = s.Get(typeof(SemanticGraphCoreAnnotations.BasicDependenciesAnnotation));
SemanticGraph enhanced = s.Get(typeof(SemanticGraphCoreAnnotations.EnhancedDependenciesAnnotation));
if (enhanced == null)
{
enhanced = s.Get(typeof(SemanticGraphCoreAnnotations.BasicDependenciesAnnotation));
}
int beginIdx = headword.Index() - 1;
int endIdx = headword.Index();
// handle "you all", "they both" etc
if (sent.Count > headword.Index() && sent[headword.Index()].Word().Matches("all|both"))
{
IndexedWord c = dep.GetNodeByIndex(headword.Index() + 1);
SemanticGraphEdge edge = dep.GetEdge(headword, c);
if (edge != null)
{
endIdx++;
}
}
IntPair mSpan = new IntPair(beginIdx, endIdx);
if (!mentionSpanSet.Contains(mSpan) && (!InsideNE(mSpan, namedEntitySpanSet)))
{
int dummyMentionId = -1;
Mention m = new Mention(dummyMentionId, beginIdx, endIdx, sent, basic, enhanced, new List <CoreLabel>(sent.SubList(beginIdx, endIdx)));
m.headIndex = headword.Index() - 1;
m.headWord = sent[m.headIndex];
m.headString = m.headWord.Word().ToLower(Locale.English);
mentions.Add(m);
mentionSpanSet.Add(mSpan);
}
// when pronoun is a part of conjunction (e.g., you and I)
ICollection <IndexedWord> conjChildren = dep.GetChildrenWithReln(headword, UniversalEnglishGrammaticalRelations.Conjunct);
if (conjChildren.Count > 0)
{
IntPair npSpan = GetNPSpan(headword, dep, sent);
beginIdx = npSpan.Get(0);
endIdx = npSpan.Get(1) + 1;
if (",".Equals(sent[endIdx - 1].Word()))
{
endIdx--;
}
// try not to have span that ends with ,
AddMention(beginIdx, endIdx, headword, mentions, mentionSpanSet, namedEntitySpanSet, sent, basic, enhanced);
}
}
/// <summary>
/// Returns true if
/// <paramref name="word"/>
/// has an infinitival "to" attached to it.
/// </summary>
private static bool HasTo(SemanticGraph sg, IndexedWord word)
{
/* Check for infinitival to. */
if (sg.HasChildWithReln(word, UniversalEnglishGrammaticalRelations.Marker))
{
foreach (IndexedWord marker in sg.GetChildrenWithReln(word, UniversalEnglishGrammaticalRelations.Marker))
{
if (Sharpen.Runtime.EqualsIgnoreCase(marker.Value(), "to"))
{
return(true);
}
}
}
return(false);
}
/// <summary>
/// Returns true if
/// <paramref name="word"/>
/// has an inflection of "be" as an auxiliary.
/// </summary>
private static bool HasBeAux(SemanticGraph sg, IndexedWord word)
{
foreach (IndexedWord aux in sg.GetChildrenWithReln(word, UniversalEnglishGrammaticalRelations.AuxModifier))
{
if (aux.Value().Matches(BeRegex))
{
return(true);
}
}
/* Check if head of conjunction has an auxiliary in case the word is part of a conjunction */
IndexedWord gov = sg.GetParent(word);
if (gov != null)
{
SemanticGraphEdge edge = sg.GetEdge(gov, word);
if (UniversalEnglishGrammaticalRelations.Conjunct.IsAncestor(edge.GetRelation()))
{
return(HasBeAux(sg, gov));
}
}
return(false);
}
private void ExtractMentionForHeadword(IndexedWord headword, SemanticGraph dep, ICoreMap s, IList <Mention> mentions, ICollection <IntPair> mentionSpanSet, ICollection <IntPair> namedEntitySpanSet)
{
IList <CoreLabel> sent = s.Get(typeof(CoreAnnotations.TokensAnnotation));
SemanticGraph basic = s.Get(typeof(SemanticGraphCoreAnnotations.BasicDependenciesAnnotation));
SemanticGraph enhanced = s.Get(typeof(SemanticGraphCoreAnnotations.EnhancedDependenciesAnnotation));
if (enhanced == null)
{
enhanced = s.Get(typeof(SemanticGraphCoreAnnotations.BasicDependenciesAnnotation));
}
// pronoun
if (headword.Tag().StartsWith("PRP"))
{
ExtractPronounForHeadword(headword, dep, s, mentions, mentionSpanSet, namedEntitySpanSet);
return;
}
// add NP mention
IntPair npSpan = GetNPSpan(headword, dep, sent);
int beginIdx = npSpan.Get(0);
int endIdx = npSpan.Get(1) + 1;
if (",".Equals(sent[endIdx - 1].Word()))
{
endIdx--;
}
// try not to have span that ends with ,
if ("IN".Equals(sent[beginIdx].Tag()))
{
beginIdx++;
}
// try to remove first IN.
AddMention(beginIdx, endIdx, headword, mentions, mentionSpanSet, namedEntitySpanSet, sent, basic, enhanced);
//
// extract the first element in conjunction (A and B -> extract A here "A and B", "B" will be extracted above)
//
// to make sure we find the first conjunction
ICollection <IndexedWord> conjChildren = dep.GetChildrenWithReln(headword, UniversalEnglishGrammaticalRelations.Conjunct);
if (conjChildren.Count > 0)
{
IndexedWord conjChild = dep.GetChildWithReln(headword, UniversalEnglishGrammaticalRelations.Conjunct);
foreach (IndexedWord c in conjChildren)
{
if (c.Index() < conjChild.Index())
{
conjChild = c;
}
}
IndexedWord left = SemanticGraphUtils.LeftMostChildVertice(conjChild, dep);
for (int endIdxFirstElement = left.Index() - 1; endIdxFirstElement > beginIdx; endIdxFirstElement--)
{
if (!sent[endIdxFirstElement - 1].Tag().Matches("CC|,"))
{
if (headword.Index() - 1 < endIdxFirstElement)
{
AddMention(beginIdx, endIdxFirstElement, headword, mentions, mentionSpanSet, namedEntitySpanSet, sent, basic, enhanced);
}
break;
}
}
}
}