public List<List<Mention>> extractGoldMentions(CoNLL2011DocumentReader.Document conllDoc) {
List<CoreMap> sentences =
conllDoc.getAnnotation().get(CoreAnnotations.SentencesAnnotation.class);
List<List<Mention>> allGoldMentions = new ArrayList<List<Mention>>();
CollectionValuedMap<String, CoreMap> corefChainMap = conllDoc.getCorefChainMap();
for (int i = 0; i < sentences.size(); i++) {
allGoldMentions.add(new ArrayList<Mention>());
}
int maxCorefClusterId = -1;
for (String corefIdStr : corefChainMap.keySet()) {
int id = Integer.parseInt(corefIdStr);
if (id > maxCorefClusterId) {
maxCorefClusterId = id;
}
}
int newMentionID = maxCorefClusterId + 1;
for (String corefIdStr : corefChainMap.keySet()) {
int id = Integer.parseInt(corefIdStr);
int clusterMentionCnt = 0;
for (CoreMap m : corefChainMap.get(corefIdStr)) {
clusterMentionCnt++;
Mention mention = new Mention();
mention.goldCorefClusterID = id;
if (clusterMentionCnt == 1) {
// First mention in cluster
mention.mentionID = id;
mention.originalRef = -1;
} else {
mention.mentionID = newMentionID;
mention.originalRef = id;
newMentionID++;
}
if (maxID < mention.mentionID) maxID = mention.mentionID;
int sentIndex = m.get(CoreAnnotations.SentenceIndexAnnotation.class);
CoreMap sent = sentences.get(sentIndex);
mention.startIndex =
m.get(CoreAnnotations.TokenBeginAnnotation.class)
- sent.get(CoreAnnotations.TokenBeginAnnotation.class);
mention.endIndex =
m.get(CoreAnnotations.TokenEndAnnotation.class)
- sent.get(CoreAnnotations.TokenBeginAnnotation.class);
// will be set by arrange
mention.originalSpan = m.get(CoreAnnotations.TokensAnnotation.class);
// Mention dependency is collapsed dependency for sentence
mention.dependency =
sentences
.get(sentIndex)
.get(SemanticGraphCoreAnnotations.CollapsedDependenciesAnnotation.class);
allGoldMentions.get(sentIndex).add(mention);
}
}
return allGoldMentions;
}
public static List<List<Mention>> makeCopy(List<List<Mention>> mentions) {
List<List<Mention>> copy = new ArrayList<List<Mention>>(mentions.size());
for (List<Mention> sm : mentions) {
List<Mention> sm2 = new ArrayList<Mention>(sm.size());
for (Mention m : sm) {
Mention m2 = new Mention();
m2.goldCorefClusterID = m.goldCorefClusterID;
m2.mentionID = m.mentionID;
m2.startIndex = m.startIndex;
m2.endIndex = m.endIndex;
m2.originalSpan = m.originalSpan;
m2.dependency = m.dependency;
sm2.add(m2);
}
copy.add(sm2);
}
return copy;
}
